DATACENTREHUB Issue 1
Are Modular Data Centres the Lego of IT? ➲ Business Continuity Planning ➲ The World’s First Climate Positive Data Centre ➲ Data Protection In The Modern Data Centre
Data Centre Hub March 2015 | 1
SAY GOODBYE TO DOWNTIME THANKS TO DUAL SIDED LOCKING POWER CABLES!
The zLock dual sided locking cables simply replace existing power cables. All connected IT devices are well protected against accidental and vibration disconnects and the resulting power supply disruptions, downtime and equipment damage. • Unique power cable that locks both ends • Locks automatically (C13) and via twist lock (C14) • Requires no mating plug or receptacle • Prevents power cables from being dislodged • Increases the reliability of power distribution • Available in C13 to C14 and C19 to C20 • Choice of different colours and lengths
For a full range of power solutions contact info.uk@daxten.com, +44 (0)20 8991 6200 or visit www.daxten.com/uk/
2 | Data Centre Hub March 2015
®
| Contents
Quick Look Business Continuity Business Continuity Planning
10
Case Studies Power monitoring advantages
22
Cutting the Cost of UPS Technology
26
2bm Double Award Winners for Data Centre design in North America
7
8
Converged Infrastructure Flash Architecture Taking A Holistic View On Networks
12 36
Data Centres Data Protection In The Modern Data Centre
8
Are Modular Data Centres the Lego of IT?
18
The world’s first climate positive data centre is built in Sweden
34
14
22
DCIM Managing Power and Cooling with DCIM Software has Come of Age!!! 30
Regulars News
5
Power The reliability of power in the UK
14
Security Securing 3rd Party Access
6
10 Data Centre Hub March 2015 | 3
Foreward |
Publisher & Managing Director: Peter Herbert
Design: LGN Media
The views expressed in the articles and technical papers are those of the authors and are not endorsed by the publishers. The author and publisher, and its officers and employees, do not accept any liability for any errors that may have occurred, or for any reliance on their contents. All trademarks and brandnames are respected within our publication. However, the publishers accept no responsibility for any inadvertent misuse that may occur. This publication is protected by copyright Š 2015 and accordingly must not be reproduced in any medium. All rights reserved. Data Centre Hub stories, news, know-how? Please submit to peter.herbert@datacentrehub.com
4 | Data Centre Hub March 2015
Welcome to Data Centre Hub the new bi-monthly online magazine that is aimed at serving the data centre industry with all the latest news and developments. The data centre market is a fast moving industry with new products being brought to market to make data centres operate more efficiently, and continue to adapt as technology changes in the coming years. Data Centre Hub is therefore a platform for all data centre managers to keep informed of what is happening within the market. This issue will cover security, data protection, business continuity, power and the growing market of modular data centres. Data Centre Hub in collaboration with LGN Media have also launched a series of Summits with the first event taking place on 30th September 2015 at Manchester United’s football stadium. The Summit is a one day event and will focus on topics related to the north of England within the data centre market. Further information regarding the Manchester Summit can be found in this issue on the back page I hope that you enjoy the magazine and look forward to receiving your views and comments in the coming months. Peter Herbert
| News
85% of enterprises say cloud has lived up to the hype More compute and storage in the cloud than on-premises by 2024 reveals Tata Communications’ global survey London – 26th February 2015 – Research conducted by Tata Communications, a leading provider of A New World of Communications™, has found that organisations with 500+ employees are seeing tangible benefits from cloud computing. 85% say cloud had lived up to industry hype, with 23% declaring that cloud had exceeded their expectations. According to the global survey, this groundswell of positive opinion has been contributed to by enterprises experiencing tangible benefits when using cloud services – with 83% of enterprises feeling they have experienced benefits they did not expect to see. The most popular of these benefits cited by respondents were: increased productivity (69%), better access to data (65%), and reductions in costs (63%). The findings also showed that by 2024 off-premises storage will have overtaken on-premises alternatives, with enterprises forecast on average to have 58% of their compute and data storage held in the cloud in ten years’ time compared with 28% currently. When asked how much of their IT provision would be stored in private cloud specifically by 2024, respondents predicted an average of 52% - highlighting the dominance of private cloud storage. However, 94% of respondents say they would be more likely to adopt a hybrid cloud computing model if connections with the public cloud were more predictable. Furthermore, the research suggests that enterprises are realizing competitive advantages from cloud deployments. Nearly two thirds (65%) of respondents said using the cloud had led to increased speed of access to technology. In terms of streamlining business
processes, over two thirds (67%) have experienced reduced delivery times to clients and partners. 54% have seen the delivery time of new services to new markets and / or geographies reduced. Julie Woods-Moss, CMO and CEO of NextGen Business at Tata Communications, says “This independent research shows that the cloud has exited the hype cycle and entered the real adoption phase for businesses globally. It is now a strategic investment and a competitive differentiator. As a result, the network that enables cloud adoption is critical, and new services, such as our IZO™ network platform and partner ecosystem, are fundamental to creating a simple way for companies to adopt the cloud model most suited to business requirements – be that private, public or hybrid.” Today, over 24% of the world’s internet routes travel over Tata Communications’ network, which is the largest wholly-owned subsea cable network in the world. Its Tier 1 IP network provides backbone connectivity to over 240 countries and territories across 400 points-ofpresence, as well as nearly 1 million square feet of data centre and collocation space worldwide. Tata Communications’ IZO™ network platform ecosystem currently consists of over 20 service providers covering countries accounting for 85% of the world’s GDP. It also covers three of the largest cloud platforms, including Microsoft Azure and Google Cloud Platform, with over 50 data centres across the globe. The IZO™ network delivers the most comprehensive cloud enablement platform available today. To view the research video, please visit http://youtu.be/_ DJSizMWYDk To read the complete report, please visit www. tatacommunications.com/cloud
All the Latest Data Centre Hub News
Geist DCIM in Top Ten of CRN Data Centre 100 List February 16, 2015. Lincoln, NE. – Geist, a leading provider of data centre infrastructure solutions including PDUs, monitoring devices, sophisticated cooling and DCIM, has been named in the top ten of The Channel Company’s annual CRN Data Centre 100 List in the USA for the third consecutive year. The CRN Data Centre 100 List is devised by CRN (Channel Reseller News) editorial staff based on a combination of extensive research and nominations from solutions providers. CRN highlights providers that represent their respective fields with a combination of innovation, market share, market presence, buzz, technology capabilities and channel capabilities. This marks the start of a year during which Geist plans to unveil a series of new DCIM products for every environment: from a few racks to a complete enterprise solution. Featured in the CRN Data Centre 100 List, Matt Lane, Geist DCIM President said: “We are delighted to appear again in CRN report, and we look forward to adding even more value to our products throughout 2015. We are listening to our customers and are driving forward diverse solutions to address their needs and concerns. Our focus is firmly on improving our DCIM tool sets to add value to businesses from day one of implementation.” Geist’s Environet Facility DCIM solution was recently boosted by the introduction of Environet Asset; a complete Asset Management tool that is generating considerable interest in the Data Centre marketplace. That will soon be followed by the introduction of a low-cost, easy-to-deploy solution for businesses looking to install essential DCIM on a budget. www.geistglobal.com
Data Centre Hub March 2015 | 5
News |
Data centre supplier Minkels appoints Christiaan van Terheijden as its new CEO Veghel, Netherlands, 5 February 2015 - Minkels, a worldwide supplier of data centre solutions and part of the quoted company Legrand (Euronext Paris: LR), has appointed Christiaan van Terheijden (38) as the company’s new Chief Executive Officer (CEO). Christiaan van Terheijden has the task of strengthening the global positioning of the Minkels brand by rolling out the business in new export countries. Mr van Terheijden has also been appointed with the aim of consolidating the existing business and creating further synergy between Minkels and Legrand. Minkels specialises in data centres, a manufacturer that has accumulated many years of knowledge within the international data centre sector, offering modular and integrated total solutions for the development of corporate and commercial data centres. The Legrand parent organisation is a worldwide specialist in electrical intelligence and digital infrastructures for buildings. Christiaan van Terheijden has been brought in to strengthen the global position of the Minkels brand in the data centre market, to consolidate existing business and to create further synergy between Minkels and Legrand. Mr van Terheijdens previous roles include a variety of international functions at the parent company Legrand. His most recent function was as the Commercial Director for Legrand Group in Austria and Germany, where he managed four branches and the Vienna head office. Before that he was Export Manager for many years, both for the Legrand Group and for the cable management specialist Van Geel, which like Minkels was also acquired by Legrand a few years ago. Combining forces Minkels is one of the leading players in the data centre market, said Mr. van Terheijden. It is a strong and sturdy brand with innovative solutions, a brand that we would very much like to expand further internationally. The infrastructure solutions that
Legrand installs in buildings are very interesting for Minkels customers. This offers a significant extension to how well we can meet their data centre requirements, with innovative and proven technologies from a worldwide specialist in electrical intelligence. On top of that, both Minkels and Legrand offer solutions that are extremely energy-efficient. At the same time, expanding the Minkels portfolio with solutions from Legrand that are targeted at data centres will allow Minkels to get involved in new data centre projects at an earlier stage. Legrand carries out lots of electrical and technical installation projects,explained Mr van Terheijden. This means that Minkels can have access to the tendering process earlier than it currently does, from the moment that the basic construction of a data centre starts. I anticipate this giving a significant boost to further turnover growth at Minkels. International expansion Mr van Terheijden’s knowledge and expertise in exporting and global marketing approaches within the Legrand Group will also contribute to further international expansion of Minkels global key accounts. He adds, ‘Minkels is the centre of excellence for data centres within Legrand and has a strategic leading position for rolling out data centre concepts in the market. Mr van Terheijden also expects that intensifying the cooperation between Minkels and the branches and production units of Legrand throughout the world will make a positive contribution to the currently adopted strategy of a global presence with local excellence. I know the Legrand organisation pretty well now, having visited all the Legrand countries myself in my former role as Export Manager. So its not just a question of knowing the German and Austrian markets very well the links and contacts with other Legrand companies are present too. Intensifying the interactions in terms of sales and knowledge transfer between Minkels and Legrand will make sure that we are able to break open new markets for Minkels.
Belden Improves Network Intelligence and Flexibility with RiT Technology’s IIM Solution Ridgewood, NJ – Febuary 10, 2015 – RiT Technologies (NASDAQ: RITT), a leading provider of Intelligent infrastructure Management (IIM) and cabling solutions announced today that Belden, a global leader in signal transmission solutions for mission-critical applications, has enhanced its network infrastructure management product portfolio in the Asia Pacific (APAC) region with RiT’s IIM solutions. RiT Technology’s IIM solution provides complete real-time visibility and control of all network physical components including status of devices and their location, in both inter-connect and crossconnect configurations. This capability enables IT departments to quickly identify connectivity issues, points of failure, and unused or underutilized equipment for improved efficiency and uptime. “The rising demand for bandwidth, 6 | Data Centre Hub March 2015
storage and computing power with limited space and resources is making the management of IT infrastructure more complex,” says Rajesh Shenoy, Director Business Development at Belden. “IT departments need complete visibility, control and management of all network components and their connections to improve their operational efficiency, minimize risk of downtime and effectively plan for the future.” “By leveraging Belden’s market and industry leadership, our solution will bring network intelligence and independence to healthcare, financial and government markets in APAC,” said Assaf Skolnik, VP Sales EMEA and APAC at RiT Technologies. “Our partnership has already resulted in several customers and we believe this is only the beginning of our relationship.”
About Belden Belden Inc., a global leader in high-quality, end-to-end signal transmission solutions, delivers a comprehensive product portfolio designed to meet the mission-critical network infrastructure needs of industrial, enterprise and broadcast markets. With innovative solutions targeted at reliable and secure transmission of rapidly growing amounts of data, audio and video needed for today’s applications, Belden is at the center of the global transformation to a connected world. Founded in 1902, the company is headquartered in St. Louis and has manufacturing capabilities in North and South America, Europe and Asia. For more information, visit us at www.belden. com; follow us on Twitter @BeldenInc
News Case | |Study
2bm Double Award Winners for Data Centre design in North America The NAHPC facility was a finalist in three award categories for its cutting edge design and outstanding delivery team Earlier this year 2bm Ltd (2bm) in partnership with Digital Realty Trust (DLR) delivered a 2MW class leading, low energy, high performance computing (HPC) data centre in Austin Texas for their client ARM. The NAHPC facility was a finalist in three award categories for its cutting edge design and outstanding delivery team. This month ARM, DLR & 2bm scooped up two of the awards at the DatacenterDynamics North America Awards 2014 in New York. These were in the ‘Green Data Centre’ and ‘Innovation in the Medium-Data Centre’ categories; narrowly missing out on a hat trick for a third award for ‘Special Assignment Team’ of the year. ARM, the FTSE 100 Company at the heart of the world’s most advanced digital products, required a low-energy dedicated data centre that aligned with their ethos of innovation, efficiency and sustainability. 2bm was therefore tasked with the design of multiple, innovative, energy saving features within a single
facility. In partnership with DLR the delivery of the completed facility was acknowledged by all as a comprehensive success. ‘We had an immediate need for a high efficiency, high density data center in Austin, Texas. But more importantly, we have a longterm need for a partner who can potentially help us with our go forward data center strategy.’ — John Goodenough, Vice President of Design Technology and Automation at ARM Holdings Located in Austin TX, the project marks the first time 2bm’s Design & Build Division has been involved with a project in North America. Austin’s climate posed a real challenge in providing a low energy, sustainable, zero water consumptive data centre. From the start of the design process to handover, the facility was completed within a period of just 10 months. In addition, the scale of the project (in excess of $22m) along with the international makeup of the design and delivery team makes
2bm was therefore tasked with the design of multiple, innovative, energy saving features
the accomplishment even more noteworthy. The opportunity arose as a result of 2bm and ARM’s previous collaboration in building the awardwinning data centre in Cambridge, UK, which was awarded Gold CEEDA Certification. The successes seen at this site provided the catalyst for the design and build of the new facility in Austin which subsequently became the first data centre worldwide to be awarded Gold CEEDA Design-Operated Certification. The design of the ARM NAHPC data centre fully embraces and surpasses the sustainability credentials and the requirements of the EU code of conduct for data centres and is arguably the most efficient HPC data centre in Texas. The facility is an exceptional example of an energy efficient data centre with every aspect meeting and in some cases surpassing the requirements of the EU code of conduct for data centres. We believe that the design of the ARM NAHPC data centre fully embraces and surpasses the sustainability credentials required for commercial buildings with the city of Austin.’ — CEEDA Recognition The two awards now join 2bm’s fast-growing repertoire of successes and further demonstrates its ability to design and deliver high quality, energy efficient data centres worldwide. Data Centre Hub March 2015 | 7
security |
Securing 3rd Party Access
By Stuart Facey, VP International, Bomgar
The challenge of security and easy access. Stuart Facey discusses the challenge of thirdparty access security issues... Challenges Donald Rumsfeld provided a famous quote regarding problems of which we are aware: “There are known unknowns. That is to say there are things that we now know we don’t know. But there are also unknown unknowns. There are things we do not know and cannot know. So … we do the best we can and we pull all this information together.” In IT security, there are constant challenges, such as hackers, insider threats and malware. However, these fall into the bucket of ‘known unknowns,’ in that you should be aware of the general risks and have taken steps to prevent them. Another ‘known unknown’ that often doesn’t get the attention that it should is third party access. For many reasons, companies rely on external suppliers to provide them with IT services and support, which requires these vendors to access a company’s systems, typically from a remote location. According to research by analyst firm Ovum, in Western Europe 88 per cent of companies allowed remote access to their networks by outside suppliers. While the majority of those surveyed had only a handful of companies with access, companies with tens or hundreds of outsourcing partners were also found. Support These third parties require access to your IT network in order to provide that support. However, many IT teams don’t have any insight into what their third party suppliers are doing once they are inside the network. So while they may know 8 | Data Centre Hub March 2015
which third party organisations have access, exactly who is accessing their systems, when and what they’re doing is commonly unknown. Being able to manage, track and audit access by third parties should be viewed as critical to maintain security and compliance. As many have said before, you can’t `outsource’ security, so you can’t rely on your vendors to keep your network secure for you. The biggest example of this so far has been US retailer Target – the company had more than 40 million credit card records stolen, and the initial attack came through a third party supplier. With this in mind, there are some simple steps that companies can take in order to improve third party access security: 1. Carry out an audit The key question to ask is how many third party relationships you have, and whether any of these companies require access to your IT systems remotely. Following the principle of least privilege, give these companies access to only the systems on which they need to work, and consider limiting what days and times they have access, further deterring unsanctioned behaviour. 2. Consider your remote access tools Once you have a list of vendors with valid remote access requirements, it’s important to look at how those vendors are currently interacting with your network. Are they using their own remote access tool of choice, or one that you own? How
does it gain access to your network and where are the audit logs stored? When possible, require all third parties to use a consolidated remote access tool you own or manage so you’re in complete control and have a comprehensive record of who accesses your systems and when. Many remote access tools rely on open ports on the firewall, specifically the approach that free remote access tools use in order to connect across the Internet. While they might be useful for small business or personal use, open firewall ports like RDP port 3389 are favourites for hackers and can be dangerous to run. If you do decide to use tools like this, be aware of the risks and how to mitigate them. 3. Review the audit trail Regularly review what your vendors are doing while they are in your systems and set up alerts for any unexpected activity, such as a vendor logging in outside normal working hours. Conclusions Overall, third party access can be a big challenge to consider, especially when it comes to the role-based access and compliance requirements that are contained in regulations like the PCI card payments framework for data security. As more outsourcing takes place, companies will have to consider their approach to managing security and remote access. To get the most from these relationships, security and vendor access does have to be considered from the start. www.bomgar.com
Data Centre Hub March 2015 | 9
data | centres
Data Protection In The Modern Data Centre Why data is king in the modern data centre. Sheldon D’Paiva explains how the data centre ecosphere is evolving...
By Sheldon D’Paiva, Product and Solutions Marketing, Nimble Storage
Storage silos Storage silos have long ruled the traditional data centre. Businesses have been purchasing separate primary and secondary storage systems – and for good reason. The end users of business critical applications such as email and databases demand the performance that primary storage systems deliver. When backing up those applications however, the backup data does not need to be accessed in real-time, and the decision point often shifts back to cost, at which secondary storage systems excel. But there is an unrelenting tide eroding those dividing walls between silos. Nimble Storage conducted a survey on data protection with 1,600 participants, which found the majority of enterprises believe they cannot afford to lose more than 6 hours-worth of data (RPO), and that they must be able to recover protected data in less than 6 hours (RTO). Using separate primary and secondary storage systems mandates that data be read from the primary system, moved across 10 | Data Centre Hub March 2015
the network, and then written to the secondary storage system – making 6-hour RPOs and RTOs all but impossible. Meeting strict data protection requirements at scale requires a different approach – one that doesn’t require the traditional read-move-write methodology that impacts production infrastructure. High availability To deliver high data availability, a storage system must be built on a fault tolerant architecture. A fault tolerant architecture means that the system should be designed to tolerate failures at multiple levels, and is essential in order to deliver “five nines” availability – or system availability of 99.999 per cent. Failures at the component level must be detected and corrected. For example, RAID technologies use redundancy to recover from drive failures at a component level. A system should also be designed to be fault tolerant at a sub-system level to eliminate a single point of failure. Enterpriseclass storage systems are typically designed with redundant controllers
that can take over in the event of a failure. When it comes to data protection, storage snapshots provide recovery point objectives (RPOs) and recovery time objectives (RTOs) unmatched by traditional methods that have a read-movewrite impact on production infrastructure. Storage snapshots are essentially a point-in-time version of the data on disk. For effective use in aggressive data protection scenarios, they need to be very efficient – with a well thought out implementation and metadata (pointers to the data) layer, so the data itself is not moved or copied every time a snapshot is taken, and only the changed blocks are stored. For very aggressive data protection needs, the storage system must be able to take snapshots every 15 minutes (15 min RPOs) and also be able to retain those snapshots cost-effectively. Leveraging SSDs Leveraging SSDs for snapshot metadata and high-density disk for snapshot data enables a storage
data | centres
system to take frequent snapshots without impacting the performance of critical workloads, and also store the snapshot data cost-effectively to address the majority of data recovery cases. Finally, storage snapshots can also be used to recover data even in the event of a complete array failure or site outage, by replicating the snapshots to another storage system – which is often at a remote site for disaster recovery purposes. A fault tolerant architecture is the baseline for high data availability, and an efficient storage snapshot implementation can deliver on the service level agreements required for aggressive data protection, but data analytics can contribute to pro-actively ensuring the overall wellness of the storage environment. In fact, data analytics can improve data availability beyond the “five nines� level. In order to do this though, data analytics need to be well integrated within the storage system. The
To deliver high data availability, a storage system must be built on afault tolerant architecture storage environment must also be monitored for various metric values that number in the millions for a single system - from physical metrics such as fan speeds and power supplies to data services metrics such as volume sizes and performance levels. Although the storage system needs to be able to collect and track a vast number of metric values, processing them all on the storage system itself would put an additional load on the system that could impact performance. A cloud-deployed service can collect the analytics telemetry data from the storage system and then processes it in the cloud to offload the heavy lifting. It also removes the need to deploy additional
infrastructure onsite, while enabling the ability to monitor multiple systems from anywhere to ensure data availability and data protection compliance. A fault tolerant architecture, modern storage snapshots, and powerful data analytics form the basis for a holistic approach to backup and data availability, resulting in storage systems that operate in peak condition at all times. These three key underpinnings can provide both high data availability, as well as an effective data protection platform for aggressive requirements. www.nimblestorage.com
Data Centre Hub March 2015 | 11
Business | Continuity
Business Continuity Planning The best approach to data recovery. Russell Cook makes clear the best approach to data recovery
By Russell Cook, Managing Director at SIRE Technology
The advent of the cloud offers significant benefits in terms of Business Continuity of Service. 12 | Data Centre Hub March 2015
Business | Continuity
Introduction When it comes to preparing for the unthinkable many companies overlook continuity planning, which should be an integral part of their daily management procedures. Even though the cloud is now a recognised and familiar term, research suggests that nearly 50 per cent of businesses do not have a continuity plan and run the risk of extended interruption. Of those businesses without a plan: 90 % that lose data will shut within two years 80 % will fail within 13 months 53 % never recoup their losses 43 % will never re-open after the disaster!
Technology in the shape of communication mediums and CRM systems mean less paperwork is generated, but as more and more business-critical information is held digitally, businesses are more exposed in the event of a disaster. While most businesses take the threat of data loss seriously and put in place backup procedures, it is the speed and quality of the recovery process that’s critical. Solutions that offer damage limitation are not sufficient. You need to know that in the event of any incident, all your business applications and data will be instantly and completely available to you, ensuring you can continue trading with minimum disruption.
What makes your business tick? A wide range of assets come together to create the entity that is your business and all must be safeguarded against any number of hypothetical eventualities. The tangible assets that are vital to your operations are relatively easy to protect. What about the truly vital assets, without which your business couldn’t even function? The communications, transactions, contracts and records that represent the results of years of trading - can they be replaced? What about the applications required to support and manage all of this accumulated knowledge and intelligence?
Business Continuity Benefits The advent of the cloud offers significant benefits in terms of Business Continuity as a Service (BCaaS), where the system you select will mirror your server to the cloud. This offers up to 38 per cent reduction in infrastructure cost, provides protection for your business and is accessible at a flick of a switch and with instant visualisation. Another benefit, that’s often either forgotten or not fully appreciated, is the systems ability to de-duplicate data. Much of the data on your organisation’s live system will be copied time and time again. For example, when you cc an email
• • • •
to other people in the business the same data is saved multiple times across the business. With a modern disaster recovery (DR) system only one version of the email will be stored. At its most effective this de-duplication system can deliver a staggering reduction in data storage of up to 95 per cent! So, disaster has struck and it is your busiest time of the year, what do you need from your business continuity support? Hopefully the system you have in place has a simplified backup administration that delivers a scalable solution, encompassing the virtual environments, databases, email and enterprise resource planning. These various elements are best delivered via the following components: •
•
•
Technology in the shape of communication mediums and CRM systems mean less paperwork is generated, but as more and more business critical information is held digitally,
Image-based back up efficiency; data is converted directly to virtual machines. Since each backup is a fully bootable virtual machine there is no need for a conversion to occur before performing a restore. Data is always available as there are no complicated rollup or restore processes. Instant virtualisation in just seconds; backups can be virtualised, either locally or on an appliance such as Reviiver, or in the remote cloud. CPU, memory and network resources can be dynamically configured allowing changes to be made without restarting the virtual machine. Centralised management accessible; if using Reviiver all the appliances can be controlled through a central web interface that provides access to critical statistics and remote management to remote servers and virtual machines through a secure VPN connection.
Conclusion We all hope that we won’t need to call on the insurances we take out, but taking the time to plan for the unthinkable will bring you a lot more than just peace of mind. So, as you draw up your objectives for 2015 make sure that your business continuity planning is right at the bottom of the list again. www.sire.co.uk
Data Centre Hub March 2015 | 13
converged | Infrastructure
Flash Architecture Evaluating storage for applications. How flash architectures have changed the storage market. Introduction Flash leverages solid-state technology to store and retrieve data making it significantly faster than hard disk, which is especially beneficial in virtualised environments where data access is highly random. However, flash technology has limitations. On average, it tends to be between 8 and 20 times more expensive than hard disk. It also suffers from wear leveling: a particular cell in flash can only be ‘written to’ a certain number of times (also referred to as a P/E cycle) before it becomes invalid. Flash can be deployed on both servers and storage to accelerate data access. The right approach is determined by the application’s particular performance, capacity, and data protection needs. Flash on Storage This can be divided into: (1) Old-school architectural models that retrofit flash over hard disk architecture, and (2) Newer architectures designed for flash. Flash offers improvements in both cases. With model (1) above, flash is used as cache or tier. In the caching model, read accesses are accelerated, but writes to storage are not. The tiering model uses flash as an intermediate tier for storing hot data (while cold data resides on disk). In general, this model works well in instances where the data being accessed is on flash. When it is not, data must be ‘promoted’ from disk to cache while other data on the cache is ‘demoted.’ That requires significant system overhead since the promotion process is not triggered until the number of cache misses reaches a certain threshold. Only then is data promoted, and in large chunks that 14 | Data Centre Hub March 2015
increase the chances of cold data taking up precious flash capacity. Furthermore, in tiered models, flash is used as a staging tier. While write performance can be improved in this manner, this also leads to other overheads. Firstly the flash capacity needs to be large enough to accommodate peak load conditions. In addition, flash needs to be RAID protected to ensure data integrity. Finally, such models introduce bottlenecks when draining data to disk. Flash on server For performance-intensive applications, flash on server can be effective. It delivers very low, sub-microsecond level latencies. However, there is a limit to the amount of capacity that can be supported. It can also be challenging to use common data protection techniques like snapshots and replication. Flash on server works well for high performance applications where data availability is not as critical or is guaranteed through other mechanisms. Modern Architecture Modern architectures are optimised for flash’s unique capabilities. Again, there are two popular models: One in which flash is the only storage repository (known as all-flash arrays), and another in which flash is combined with hard disk (flash-disk arrays). All-flash arrays effectively
improve the system’s responsiveness. These systems use data reduction techniques like deduplication and compression to increase flash’s usable capacity, but they’re expensive because flash suffers from wear leveling, requiring the architecture to significantly over-provision to lengthen its life span. Also, because when flash is the only storage repository it must be RAID protected to prevent data loss from drive failures. Finally, these models require snapshots to be stored in flash, which is often unnecessary. Conclusion An intelligently architected flashdisk system delivers the same level of latencies/responsiveness as an all-flash array without the overhead. If the array allows flash quantities to be flexibly varied depending on workload, it eliminates the need for an expensive all-flash array. And, flash-disk architecture allows capacities to scale to petabytes of storage. Also, by storing a copy of data in disk at all times, there is no need for overprovisioning or RAID protection; snapshots and cold data are stored in disk where they belong. When evaluating storage for applications, the following factors play a key role in determining the right solution: Performance requirements • Capacity requirements • Data availability and data protection requirements • Cost-effectiveness. A well-designed flash architecture will ensure that customers do not have to trade-off one requirement for another. www.nimblestorage.com
Data Centre Hub March 2015 | 15
Power |
The reliability of power in the UK Alan Luscombe reports on the biggest threats to IT industry according to new UPS survey.
By Alan Luscombe, Sales and Marketing Director & UPS Ltd Introduction Regular increases in the price of energy and the demands placed on businesses through environmental legislation continue to be considered amongst the biggest threats to the UK IT industry. This is according to the results of Uninterruptible Power Supplies Limited’s (UPS Ltd) latest industry survey. The 2014 edition of company’s annual opinion poll, which questioned 2000 IT and data centre professionals across the UK, revealed that 77.51 per cent believe that rising power costs are still a ‘major concern to their business’, with over 60 per cent also conceding that increasing environmental legislation represents a ‘significant risk’ to their long-term prosperity. Downward Pressure However, when comparing the latest results to those from 2013, it would appear that the situation has begun to improve, with those responding negatively to both questions falling slightly from 12 months ago. The number of those claiming that the cost of energy is a major issue fell by over seven per cent since last year. Likewise, those concerned with environmental legislation fell by almost five per cent in the same period. “When we look at these results, it’s clear that there’s still a strong pressure to control energy costs and be more environmentally efficient, but the survey does 16 | Data Centre Hub March 2015
There’s still a strong pressure to control energy costs and be more environmentally efficient.
DATACENTRES are MATURING Mature Data Centres know that protecting their customers’ data isn’t just about being popular, living in the upmarket streets of London, wearing Tier III trainers or comparing the size of their PUE.
A mature data centre understands that high quality, exceptional service, low cost & ultimate flexibility combined with levels of security unsurpassed elsewhere is more important than boasting about the size of your PUE or your tier III label.
Don’t let childish boasts cloud your decision choose a data centre that offers maturity and puts your business needs first.
Contact MigSolv Today
0845 251 2255
migsolv.com
Data Centre Hub March 2015 | 17
Power | demonstrate that, at least for some, things are looking brighter. With the price of oil decreasing significantly and forecasters expecting further downward pressure, there is a realistic belief that energy prices are set to stabilise and, according to the ratings agency Moody’s, UK energy prices will remain close to current levels until 2020. This isn’t a view shared by everyone, but it’s still in stark contrast to more negative forecasts from a couple of years ago,” remarked David Renton, Managing Director for UPS Ltd. Spare capacity within the UK electricity network was another topic those questioned felt strongly about, with 74.64 per cent of respondents claiming that the reliability of power in the UK is going to become a major concern within the next ten years. This is a slight increase from the previous survey, where 71.93 per cent chose ‘strongly agree’ or ‘agree’ to the same question. This reflects
the latest Winter Outlook report published by the National Grid, which reported that the capacity margin for UK electricity has fallen to 4.1 per cent. In 2010 the figure was closer to 15 per cent. Despite the drop, the Department of Energy and Climate Change (DECC) has moved to dispel fears of winter blackouts. Conclusion Regardless of how energy costs and capacity develops, the survey’s findings underscore how those working in IT and data centrerelated industries are vulnerable to rising power costs and increasing environmental legislation, with electricity costs in particular making up a notable proportion of IT businesses opex. The survey also revealed the ways IT businesses are making pragmatic decisions to offset external pressures, confirming that product efficiency is now a key consideration when purchasing
The number of those claiming that the cost of energy is a ma jor issue fell by over seven per cent since last year
The reliability of power in the UK is going to become a major concern within the next ten years. 18 | Data Centre Hub March 2015
capital equipment for over 85 per cent of respondents. Moreover, two thirds (66.51 per cent) claimed ‘reducing operating costs’ was their primary driver for change, with 22.01 per cent citing ‘reducing their carbon footprint’ and 10.53 opting for ‘a desire to be more socially responsible’ as their biggest motivating factor. www.uninterruptible-powersupply-uk.co.uk co.uk
Innovative
Data Centres at the core of your business Sudlows are leading experts in data centre audit and consultancy. We specialise in the design, build and maintenance of energy efficient, sustainable data centre environments. Call +44 (0) 870 278 2787 or email hello@sudlows.com to discover more about what we can do. www.sudlows.com
Audit | Design | Build | Maintain
Data Centre Hub March 2015 | 19
data | centres
Are Modular Data Centres the Lego of IT? How modular can offer companies a flexible IT future. Matt Goulding discusses the industry shift toward modular data centres
By Matt Goulding, Managing Director, Cannon Technologies
IT has always liked things that were modular. It plays to the engineering geek inside most IT professionals. When Sun Microsystems announced Project Blackbox back in 2006, the company started a race to move from modular systems to modular data centres and a market that is estimated by Companies and Markets Research to be worth $40.41bn by 2018. Before Sun Microsystems (now Oracle) launched Project Blackbox - later known as the Sun Modular Datacenter - a number of companies, including Google, had played around with the idea of modularising the data centre. However, it was Sun that took the headlines when it launched the Sun Modular Datacentre in 2008. Sun took a 20-foot standard shipping container and built in all the racks, cabling and equipment required for it to become a data centre. All it needed was external power, chiller units and a network link. 20 | Data Centre Hub March 2015
Modular data centre The answer is that it all depends on whom you talk to. While there is no standardised description, most in the industry see it as a container based solution. However, the length of the container and even the width can vary by manufacturer and product range. Despite this, what makes it modular is that the data centres can be bolted together and even stacked on top of each other to create a larger facility. So what is driving this change? This is a complex question with a lot of different answers. Some of them are about the technology while others are about cost. There is no more powerful incentive to adopt an approach than cost. When Sun launched the modular data centre, it talked of being able to buy a data centre for less than one per cent of the price of a traditional facility. There were many reasons for this. There is often no planning
permission, no building costs or need to acquire an existing facility, as well as no decommissioning or reconfiguration costs as the business demand changed and no business rates on the use of a building. All of those reasons are as true today as they were back then but perhaps the biggest cost reason was speed. It can take months and even years to build or refurbish a facility but a modular data centre can be provisioned in just weeks from the order being placed. For service providers and data centre owners, modular is a huge opportunity. Google and Facebook today have hundreds of containers stacked together to create their data centres. When they need additional capacity, they just drop in more containers. Change is expensive One of the real challenges for the data centre is flexibility. For decades, the data centre didn’t change much. Even as we moved from
data | centres
For service providers and data centre owners, modular is a huge opportunity mainframes to mini-computers and into early rack-based solutions, the data centre was reasonably static. With the explosion of commodity computing and blade servers, the data centre became a place of constant change. For older facilities that means redesign, overhauling and updating in order to deliver the power and cooling requirements of new technologies. This is expensive. An overhaul of a data centre can cost millions of pounds and take months. During that period, no money is coming in and in a competitive
market that means the potential loss of customers. Modular data centres remove that need to lose customers and refurbish a facility. A refurbishment is simply a replacement module. Customer systems moved from one module to another. Meanwhile the under performing module can be replaced or updated as required. Dense computing Modular data centres are not just about the provision of extra capacity or for cover when refurbishing a data centre. With dense computing,
data centres have shrunk in size for many companies. Downsizing a data centre is just as expensive as adding more capacity. Power and cooling systems need to be maintained even if they are not being used. Data centre halls that are not being fully used, need to be partitioned to reduce waste and that means refurbishment costs. Using a modular data centres, companies can quickly downsize or move from older, larger, systems to smaller, more efficient ones. New opportunities One of the real benefits of modular data centres has been the ability to deploy in new areas. The football World Cup, the US Football SuperBowl, the summer and winter Olympics, the football World Cup - as well as other global events - require data centre facilities, especially for media. Data Centre Hub March 2015 | 21
data | centres
These events process large amounts of data, film, audio and run vast Internet sites to provide public information. Disaster recovery operations, oil rigs, intelligence headquarters in war zones and even major political conferences have also bought into modular data centres. They all need to manipulate very large volumes of data and modular data centres make this possible and provide an opportunity to do so securely. In 2010, the US government took a long and hard look at how it responded to major disasters. One of the failings was the inability of government departments to respond caused by a lack of data and IT facilities. The result was a document highlighting how departments should evaluate and then commission modular facilities. Highly efficient Modular data centres are engineered to customer requirements. As they can be accessed from all sides, the components are integrated 22 | Data Centre Hub March 2015
to create the most optimal configuration for power and cooling. Over time, as components alter, some of the initial integration may be lost but those losses will be offset by the power efficiencies of new generations of IT equipment. Conclusion The next generation of switches that run at 40, 100 Gbps and even faster, require a lot more power. Storage systems - even those with hundreds of Sold State Drives, also need a large amount of power. Modular data systems are capable of supporting racks of blade systems, petabytes of storage and very high bandwidth, something that traditional facilities would need 10 times the floor space to accommodate. Modular data centres such as the Cannon Technologies T4 MDC drive down costs, can be deployed wherever there is a need for compute power and are highly efficient. www.cannontech.co.uk
Using a modular data centres, companies can quickly downsize or move from older, larger, systems to smaller, more efficient ones
IT Cooling Solutions
The Whole Range of Data Center Cooling Solutions from a Single Source CyberAir 3 Room Cooling
CyberRow High Density Cooling
CyberCool 2 Chiller Units
CyberCon Modular Data Center Cooling
CyberHandler Air Handling Units
STULZ GmbH . Company Headquarters . Holsteiner Chaussee 283 . 22457 Hamburg . Germany products@stulz.com . Near you all over the world: with sixteen subsidiaries, six production sites and sales and service partners in more than 120 countries. www.stulz.com
D ATA C E N T R E C O O L I N G S O L U T I O N S S A L E S / S U P P O R T / S E R V I C E / S PA R E S STULZ UK Ltd . First Quarter . Blenheim Road . Epsom . Surrey . KT19 9QN 01372 749666 . Sales@stulz.co.uk www.stulz.com Data .Centre Hub March 2015 | 23
Case | Study
Power monitoring advantages Power monitoring goes a long way at TalkTalk. Andy Mullings explains how power monitoring can reap dividends...
By Andy Mullings, Data Centre Business Development Manager, Synapsys Solutions Introduction TalkTalk has for many years used Synapsys products for building management system monitoring and integration with a Trend Building Management System. In the service provider’s UK data centres, the technology integrates systems including UPS and DC power systems, Computer Room Air Conditioning (CRAC) units, metering, environmental monitoring, and so on. Synapsys’ new SIPp power monitoring solution now adds the capacity for accurate, real-time power monitoring, and TalkTalk is installing the SIPp systems for rack monitoring at all its data centres. Already retrofitted at the Manchester site’s first data hall with no downtime or rewiring required, last year’s expansion to a second data hall gave TalkTalk the opportunity to incorporate the SIPp devices from the outset. This particular site had a problem to address because each system used a different communications protocol to communicate with the outside world - including RS232, RS485, SNMP, Modbus and other system specific protocols. It was clear that finding a way to get all the different elements of the network communicating would make life a lot simpler. Monitoring systems Paul Ashford, a senior electrical 24 | Data Centre Hub March 2015
Screen technology helps the monitoring process engineer with TalkTalk, specified a SIPp system for the project, where it monitors both the AC and the DC power systems, communicating this information back to the Trend 963 Supervisor BMS. “It makes managing the system very straightforward by establishing a single method of communication with our BMS system, and this simplicity for the operator is the system’s biggest plus point for us. We’d otherwise be looking at
using different OEM software for each of the many systems that we use across the estate, monitoring systems including DC and UPS systems, power usage, backup generator status, CRAC systems, fire alarm systems and temperature and humidity levels - which is simply not practical,” he explained. To monitor load current and energy usage in the AC branch circuit monitoring system, a SIPp device simply plugs in to the power
Servers
Cloud
Datacentre
WELCOME TO YOUR 100% GREEN DATACENTRE
Work with a Datacentre provider with long term goals to reduce energy consumption
If you already colocate, you will need to think about expansion, more resilience and choosing a supplier of not just more space, but better space; one where security, power, cooling and fire concerns are all taken care of.
LIMI T TO F ED I 50 C RST LIEN TS
If you house your own servers, you may already be looking around the office, wondering where the safe, secure, cooled space is coming from for your future expansion, be it for greater resilience or more processing power. If you’re in the hosting business you need to think about your clients expanding demands.
Domains
Hosting
Overview > An industry standard 42u rack fully secured & lockable > APC PDUs to each cabinet > DDOS Mitigation Colocation Features > 24 hour access - 365 days a year > 24/7 Support > Instant reboot service > Up to 64 Amps power per cabinet > Redundant UPS protection > Redundant diesel generators > Advanced VESDA fire detection & FM200/IG55 Ad protection system > 10 Gbps network > Meeting, Storage & Build rooms available 100% Uptime
24/7 Support
The Netcetera Dataport
Unlimited Bandwidth
Offshore Hosting Options
19
Years 1996-2015
ISO
9001 14001 27001
From £259/pm
netcetera
www.netcetera.co.uk/dch Call us FREE on 0800 808 5450 sales@netcetera.co.uk Data Centre Hub March 2015 | 25
All prices exclude VAT at 20%
Case | Study
distribution unit at the end of each equipment rack. Each SIPp can monitor up to 72 individual power circuits using micro CTs (current transformers), 36 on each feed. It converts the information obtained into a common Trend format that can be shared with the other systems used in the estate. SIPps are also used to monitor the -48V DC load current and power circuits at rack level across the facility’s data centres, providing a very robust and reliable system. The power monitoring system ensures that all generated alarms are relayed back to TalkTalk’s central operation centre, which is managed around the clock. Alarms are assessed and escalated if necessary to ensure that the appropriate level of response is carried out. Engineers can log in remotely to get precise information, even down to individual rack level. Unusual readings can be examined to see whether they have been caused by something as straightforward as a 26 | Data Centre Hub March 2015
problem with a filter in the CRAC system, in which case schedule ‘next working day’ maintenance can be scheduled. Alternatively, if for example a UPS has gone down and a generator has been activated, an immediate response is the order of the day. “It’s all about knowing where we are,” said Ashford. “There’s a lot of detail in the feeds we can choose to monitor - we are able to access a massive amount of information. We’re now rolling out additional SIPps across all our sites in London, Birmingham and Manchester to help us monitor our equipment rack power circuits and to feed this information back to our central network operation centre in Manchester,” he added. Scalable hardware The SIPp system offers a scalable hardware-only solution, with no ongoing licence or service costs. With its user-friendly graphic ‘dashboard’ display, the system presents real-time data in an easy-
Engineers can log in remotely to get precise information
to-read format. New users can get to up and running with minimal training, as the user interface is so intuitive. “Developing our system has not been without its challenges and we’ve all been on a massive learning curve, but the Synapsys team has worked with us from the start on some pretty ground-breaking projects, bringing real innovation and developing technology which today leads the industry,” Ashford concluded. www.synapsys-solutions.com
Efficiency and technology in perfect harmony.
Today, choosing a new UPS system is about more than simply protecting your critical load, you must also consider the best way to minimise your CapEx and OpEx and ensure you are future-proofing your power protection investment against any eventuality.
Just like our elite sportsmen and women, our UPS solutions utilise the very latest technological developments to deliver class-leading efficiency levels of up to 98%*, whilst still offering you the flexibility and control you need to achieve your financial and business objectives, both now and in the future.
To find out more call or email us today: 01256 386700, sales@upspower.co.uk, www.upspower.co.uk Data Centre Hub March 2015 | 27
Case | Study
Cutting the Cost of UPS Technology There’s more to UPS than meets the eye. How West Sussex County Council trimmed 75 per cent of the cost of its UPS operations…
By Kenny Green, Technical Support Manager, UPSLolutions
The public sector - svaing money is a must 28 | Data Centre Hub March 2015
Case | Study
Introduction When West Sussex County Council took over Horsham’s County Hall North building, which was formally occupied by insurance giant Royal Sun Alliance, a shift in the building’s intended use led to significant changes in its data storage and processing requirement. In turn, this had a knock-on effect on the UPS equipment used to support the critical load and maintain 24/7 availability. Following the council’s arrival, a review of the current UPS system was commissioned to identify how the changing function of the building had enabled savings to be made. Instrumental in the review process was the council’s Carbon Management Team, which is responsible for meeting targets for reducing energy consumption and carbon emissions across the council’s 300 buildings. Once the review was complete, it was clear that the council’s critical load was significantly smaller than the existing capacity, with only relatively small increases in load expected over the next five years. This meant a new UPS, specifically tailored to the council’s actual load, could be commissioned. “The old UPS had a capacity of 200 kVA; with another 200 kVA stand alone UPS just for redundancy. In itself, this was an inefficient system compared to current technology and we didn’t need anywhere near that amount of capacity so immediately we
‘hot-swap’ modularity enables capacity to be added in cost effective incremental steps as power requirements grow recognised that we could reduce the size of the UPS, thus also reducing our power and cooling requirements,” said Nicola Winser, Carbon Management Officer for the council. Finding the right solution It was at this stage that Uninterruptible Power Supplies Ltd, a Kohler company, were contacted to provide an appropriately sized and efficient UPS system that could manage the transition between the mains and a standby generator, ensuring that all data remained accessible and unaffected by any power interruptions. The decision to work with UPSL was based on the council’s and its maintenance contractor Scottish and Southern Energy (SSE) Contracting identifying them as providing best value for money for the local authority. “After looking at the PowerWAVE 9000DPA product it was also clear it offered the best solution for our needs, at a competitive price. Our review had highlighted that our actual load was less than 100 kVA and with the 9000DPA’s modular technology, we had the flexibility we needed to expand the system in line with any future growth,” said Winser The selected PowerWAVE 9000DPA system offered the council a 100 kVA load capacity, comprising three 50 kVA modules with N+1 redundancy. Should the council’s load require it, the PowerWAVE 9000DPA’s true ‘hot-swap’ modularity enables capacity to be added in cost-effective incremental steps as power requirements grow. Additionally, the ability to ‘hotswap’ modules whilst the system is online significantly reduces mean time to repair (MTTR) and simplifies system power upgrades.
Feeling the benefit “To give you an idea of the breakdown, the old UPS running costs were £7631 a year. That’s just the power consumption. On top of that, the air conditioning was costing £2500 a year. Compare that to the new system and we are expecting power costs of £1800, with cooling costing less than £600 a year. Over a five year period, we will be saving almost £39,000 in reduced energy and air conditioning costs alone,” said Winser. It wasn’t just in cash terms that the council was able to save. The new system also offered a substantial reduction in carbon emissions, a key driver for the project. “West Sussex’s annual CO2 reduction target is 320 tonnes for the entire corporate estate and although there are 300 buildings, finding projects that deliver meaningful reductions is still a challenge. That’s why this was such an important project for us. The new UPS system was able to deliver a 37 tonne reduction, due to the UPS being 11 per cent more efficient at its expected loading, so it’s a high proportion of our annual target,” she explained. West Sussex County Council was also able to receive 80 per cent of the total purchase and installation costs through the Salix Finance scheme, funded by the DECC. Salix Finance helps to empower public sector organisations to take a lead in tackling climate change by helping to increase their energy efficiency. In the case of West Sussex County Council, only 20 per cent of its own funding was required for the project with the rest coming from Salix matched funding. About the supplier Uninterruptible Power Supplies Limited (UPSL), a Kohler company, is the UK’s leading supplier of power support solutions, backing its products with comprehensive technical, installation, commissioning and maintenance services, and with an extensive track record in public sector, financial and retail channels. www.upspower.co.uk Data Centre Hub March 2015 | 29
DCIM |
Managing Power and Cooling with DCIM Software has Come of Age!!! DCIM encompasses a range of applications that can help improve data centre operations
For the past few years, the buzz in data centre trade shows around DCIM (Data Centre Infrastructure Management) software has been substantial – much of it hype, not reality. But this market sector has been growing and maturing, and the technology around it has gone from being a “bleeding edge” utopian promise to a practical reality for many larger data centres. DCIM is a category of software, not a single piece of software Just as “office” software encompasses word processing, presentations, spreadsheets, email and other applications, DCIM encompasses a range of applications that can help improve data centre operations. Analysts peg the size of the DCIM market at around $600 million, and growing around 25% per year, meaning it’s on track to exceed a billion dollars in three years.
30 | Data Centre Hub March 2015
The two core building blocks of DCIM: DCIM Monitoring and IT Asset Management (ITAM) DCIM Monitoring is about collecting information about the real-time, current, state of your data centre – with power consumption and temperatures being two of the most important areas to monitor. Monitoring includes alerting (or alarming) of failures or potential issues, and tracking data historically, over time. It is by far the largest piece of the DCIM market comprising over half of DCIM spending in this category. IT Asset Management is the nextlargest piece; this software focuses on the details of all the equipment in a data centre. Other applications riding atop DCIM Monitoring and ITAM Once the basics are in place, you know what you have in a data centre (ITAM), and how it’s
performing (DCIM Monitoring). Now it’s time to take the next step, with capacity planning and management (space, power, cooling, cabling), or controlling the facilities side with Dynamic/ Adaptive Cooling, or controlling the power state of IT equipment, power capping, load shedding, or modelling the thermal environment using Computational Fluid Dynamics (CFD) and more. Some of these applications have more “sex appeal” than the two foundations, but if you add up the market size for every one of these, put together, you get a number significantly smaller than the ITAM market alone, much less the DCIM Monitoring market. You can buy integrated suites, or you can assemble “best of breed” modules As with many other enterprise software solutions, you can select software from one vendor that has been pre-integrated (like a
| DCIM
Swiss Army Knife) or you can select individual modules and integrate them into a solution. The trade off here is ease of use and preintegration vs. selecting the “best” module that fits your solution. While the knife blade in a Swiss Army Knife may be ideal for some uses, you may prefer a sashimi knife if you’re cutting raw fish, or a bread knife. Picking and choosing each module provides you the best tools for your requirements – but integrating multiple vendor modules may be more challenging and costly. The good news is that some vendors offer easy integration with others via various APIs. For now, there is no “industry standard” API, however. Even a single DCIM module can offer a quick return on your investment. There are many benefits of DCIM, and these vary from module to module. Research by Enterprise Management Associates (available here) quantified these benefits.
DCIM Monitoring is about collecting information about the real-time, current, state of your data centre Here are six general areas where customers find benefits; in many cases, they need to focus on just one of these in order to benefit. Broader deployment can provide even more value, but you might want to consider a step-by-step approach rather than trying to “boil the ocean:” •
Power Chain efficiency. A DCIM tool should identify various areas of inefficiency, which you can address via reloading, rebalancing, consolidating, replacing, decommissioning, controlling
•
or otherwise making changes in your Facilities and/or your IT environment. It is not unusual to find opportunities to reduce PUEs (and therefore power bills) by 15-30%. Cooling efficiency. Further power reductions can be achieved by effectively raising temperatures inside a data centre; implementing monitoring to ensure raising temperatures doesn’t get you into trouble, and identifying areas of over-cooling and under-cooling, over time. By adjusting the cooling
Data Centre Hub March 2015 | 31
DCIM |
•
•
•
•
infrastructure, or moving around heat-producing IT devices and workload, one can safely raise temperatures, and every 1 degree Celsius increase can result in a 4% reduction in power costs. Operational efficiency. DCIM can help your IT and Facilities teams coordinate effectively in managing the data centre, for example, collaborating on the optimal placement of new servers relative to existing equipment – but ensuring that there is sufficient power, space, connectivity and cooling to handle the new equipment. This results in faster, betterplanned deployments, with less labour. Imagine a deployment process, which previously took 2 weeks, now taking an hour! Automation. Monitoring power, temperature and other aspects directly and continuously obviates the need for manual readings, on a oncein-a-while basis. You have more accurate and granular data, without expending labour to take measurements. Maximizing Capacity. Cascading conservative assumptions lead to inefficient allocation of power at each level of the power chain. This results in inefficient loading, but also oodles of “stranded power” (and ditto space, cooling and cabling). Many data centres operate at 2040% of their design capacity. A DCIM solution should allow you to unlock this unused capacity, installing additional equipment in your data centre. In some cases, this means you can defer or cancel planned new buildouts, saving perhaps hundreds of millions of dollars! Higher Reliability. For some data centres, having a tool which can predict problems, identify weaknesses, and being able to fix issues quickly and effectively means you will have less costly downtime. Some operators can even quantify the business value of increased reliability.
DCIM is for all types of data centres DCIM fits “owner operator” data centres, where a single entity is responsible for operating the 32 | Data Centre Hub March 2015
DCIM can help your IT and Facilities teams coordinate effectively
building, the facilities infrastructure and the IT. But it also fits the “multi-tenant data centre” (MTDC) aka “Colocation” or “Third-Party Managed” model where one party is responsible for the facility and for providing the power and the connectivity, while the tenants (or third parties) may be responsible for the equipment. And then there are the “Cloud” datacentres, many of which operate at “hyperscale” and who compete for customers on the basis of cost, service levels and performance; DCIM tools can help on all of these accounts. If you have not yet embarked on your journey into the realm of DCIM, the question isn’t whether you should – the question should be “Where do I start?”.
DATA CENTRE SUMMIT 2015 NORTH
Data Centre Summit North is the first in a series of new one-day conference focussed events, with the first set to take place at Manchester’s Old Trafford Conference Centre on the 30th of September 2015. DCS will bring all of the industries thought leaders together in one place, with the industries leading vendors and end users. The focus of the event will be placed on education, networking and debate, It will provide an open forum for delegates to learn from the best in the business. The event will also feature an exhibit hall where the industries leading companies will all show their newest products and services, together with a networking lounge so that you can make connections with like minded business professionals. Schneider Electric White paper
To enquire about exhibiting call Peter Herbert on 07899 981123 or Ian Titchener on 01353 865403
om .c ld or
Manchester’s Old Trafford Conference Centre
ew tr en ac at .d
w w
w
30th of September 2015
DATA CENTRE SUMMIT 2015 NORTH Platinum Headline Sponsor
Data | Centres
The world’s first climate positive data centre is built in Sweden ICT (Information and Communications Technology) is emerging as a new target area in terms of energy efficiency
London, United Kingdom, 26th February 2015: In the city of Falun, Sweden, the world’s first climate positive data centre is now being built. EcoDataCenter will also be one of the 13 safest in the world, meeting the highest security requirements with world-class performance. The initiator of EcoDataCenter is the energy company Falu Energi & Vatten in collaboration with the entrepreneurial company EcoDC AB. Schneider Electric will supply technology that will deliver energy efficient solutions and products to the project. ICT (Information and Communications Technology) is emerging as a new target area in terms of energy efficiency. The ICT sector alone consumes up to 10 percent of the world’s electricity. This reflects the growing cost of powering data centres, the global backbone of the Internet and the cloud. A single data centre can easily use more electricity than a mid-sized town. A large part of this energy is never used, but simply released as heat into the atmosphere, contributing to global and local warming. There 34 | Data Centre Hub March 2015
are currently over three million data centres globally. According to several reports the combined carbon dioxide emissions from these data centres are expected to exceed the emissions from the entire airline industry in only five years. This development implies an increased need for energy, which also affects the environment to a great extent. It is also a costdriver, which requires a new way of thinking, says Børge Granli, one of the founders of EcoDC AB. Sweden has proven to be an excellent and safe location for data centres due to its cold climate, reliable and renewable energy sources, political stability and lack of natural disasters. The Cushman & Wakefield and Hurleypalmerflatt Data Centre Risk Index ranks Sweden as the third safest location for data centres in the world, after the US and UK. The world’s first climate positive data center EcoDataCenter in Falun will be connected to the local energy system, which in 2013 was praised by Global District Energy Climate Award 2013 in New York as one of
the world’s best. The excess heat from the servers and IT equipment will warm the buildings in Falun through the district heating system. In the summer time, excess steam from the local electricity plant will run the machines cooling the data centre. The collaboration between the data centre and the district heating system will ensure that the emissions are reduced to the extent that EcoDataCenter will have a negative carbon dioxide footprint over the course of a year. We are connecting the data centre to an already sustainable energy system and can make use of all the energy. Thereby we are building the very first climate positive data centre in the world, says Bengt Gustafsson, CEO of Falu Energi & Vatten. The electricity powering the data centre comes solely from renewable energy sources such as solar, wind, and water power as well as secondary biofuels. Safety combined with worldclass performance EcoDataCenter is also projected to attain the highest level of availability according to Uptime Institute
Data | Centres
than regular data centres and the integration with the district heating and cooling system means that energy is used that otherwise would have been wasted. All of this lowers the costs both for us and for our clients, says Børge Granli. The first building of three will be completed in the first quarter of 2016.
Sweden has proven to be an excellent and safe location for data centres due to its cold climate Tier IV™, something only twelve other data centres in the world have managed to attain to date. It will also have an exceptionally high performance level with a guaranteed uptime of 100 percent. Furthermore EcoDataCenter will attain the highest possible security standards. Through the cooperation with the world’s leading energy specialist Schneider Electric, EcoDataCenter will access a broad portfolio of energy efficiency solutions and products that are perfect for this project. We live in an age when more and more people are concerned about the global data centre industry’s environmental effects.
Internet penetration and the Internet of Things don’t have to happen at the expense of the climate. IT security and performance is also increasingly critical for companies and organizations. Therefore, we are tremendously proud to be a part of the creation of this pioneering data centre. EcoDataCenter is the show case for green data and shows that it is very profitable to think out of the box, says Marc Nezet, Country Manager at Schneider Electric Sweden. Green is profitable With EcoDataCenter, we have also proven that greener is less expensive. The data center is significantly more energy efficient
EcoDataCenter • Three buildings, all in all 23 250 square meters. • 18 megawatts. • Up-time: 100%. • Energy sources: Sun, wind, water and secular biofuels. • Security classification: Tier IV™ (the highest possible). The data centre will be protected by heavy concrete walls and advanced security-, control and surveillance systems. • PUE (Power Usage Effectiveness): Less than 1.15 (Tier IV™ design) • Environment-friendly design with houses certified according to LEED Platinum. • Sedum (flowering plants) on the roofs cooling them throughout the summer. • Climate: Average annual temperature 5°C. Average number of days per year with max temperature >25°C is 22 days. • Site access: 2 km to highway, 25 km to airport, 2 h train to Stockholm • Sweden has approximately 4050 percent lower power costs compared to UK electricity prices, according to statistics from Business Sweden, Nord Pool Spot and the UK Department of Energy & Climate Change. For more information, please visit the website or watch the film: www.ecodatacenter.se More information from Business Sweden: http://www.businesssweden.se/datacenters/
Data Centre Hub March 2015 | 35
converged | Infrastructure
Taking A Holistic View On Networks Converged networks need a holistic approach to meet their potential. Joseph Raccuglia explains the challenges of an open platform By Joseph Raccuglia, Director, Alcatel-Lucent
Infrastructure or management? With Enterprise networks shifting focus towards application delivery, traditional siloed network environments are failing to handle the sheer volume of data let alone the needs of a mobile workforce using multiple devices and the demands of real-time and multimedia applications. Converged networks and the automation of network management functions are offering organisations the promise of a high quality experience, reduced administration and better performance delivery, but not all solutions are made equal. With all the talk about the possibilities and potential of converged networks, it is important to remember that it is ultimately business necessity that will be driving network implementation in the enterprise. There are of course different approaches to delivering a converged network, and these are often simplified and separated into two perspectives: infrastructure and management. Making the decision to update and converge your network hardware is a big decision as it is likely to underpin the IT strategy and technological potential of a business for years to come. For an organisation that wishes to deliver high quality applications to a mobilised work force, the network 36 | Data Centre Hub March 2015
must understand devices as well as associated applications. Contextual understanding of conversations between devices and applications makes it possible for the network to optimise the user experience and network performance, while at the same time lowering capital and operating expenses. A lot has been talked about the value of Software Defined Networks (SDN) in the building of a converged and coordinated infrastructure and its ability to enable infrastructure components to automate the administration. But this is still in its infancy and most business environments will not see an end-to-end solution, rather a hybrid approach with traditional infrastructure and versions of SDN working alongside each other. Rip and replace? With converged networking, there is no need for a Rip and Replace approach. The transition can be gradual. We can already provide a lot of network automation to cope with application changes through virtualisation. The key question is how we can bring together the infrastructure pieces that may be at different levels of capability? When considering the needs of an organisation, it’s important to properly audit the capability of the entire network architecture,
to see whether it can handle the applications the business requires and whether it can deliver these applications across the network to end users’ devices. There are four key questions that need to be addressed and answered when planning a move to a converged network to enable businesses to get real value from their infrastructure. 1. Overlay networking is only as good as the underlying infrastructure. Is yours up to the job? 2. Will you have the visibility required for optimised service delivery? 3. Will you be able to customise your converged network to fulfil your requirements? 4. Will your users get the access and experience they need to benefit from an improved network? Firstly, overlay networking is only as good as the infrastructure beneath - so check it out. Some of the solutions available on the market today are not truly converged at the hardware level and simply opt for a software overlay to define storage, network and compute functions which are still based on industry standard components. Overlay networking on top of a traditional infrastructure is a way of tying together disparate
converged | Infrastructure Secondly, visibility and control are key to delivering quality service - will you have it? The difficulty with traditional networks lies in their limited awareness of the applications that are generating traffic and, conversely, the new virtual application control systems are unaware of the conditions prevailing within the network. Application Fluent Network (AFN) technology provides the solution. The goal of a converged network is to bring automation to the data centre by providing a coordinated approach to network control and enabling a centralised view on network conditions. With an Application Fluent Network (AFN), enterprises can
enjoy a network that understands devices as well as associated applications. Contextual understanding of conversations between devices and applications makes it possible for the network to optimise the user experience and network performance, while lowering capital and operating expenses. So while SDN is bridging the gap between the network world and the virtualised compute world by defining a framework that uses standardised interfaces between applications and networks, AFN technology provides a rich policy infrastructure that enables everyone to work autonomously or in a coordinated fashion. AFN also adjusts and optimises
components of a network that have different levels of capability. With this approach it is possible to deliver bits of a converged and coordinated service, which is a quick and easy way to introduce some of the benefits of converged infrastructure. But overlay networking, while meeting your provisioning requirement, will not necessarily solve the performance issues - the overlay will only ever be as good as the underlying infrastructure, and if the infrastructure is not good enough, performance will be a growing issue. Infrastructure that is built on a switching fabric that creates a ‘mesh’ network formed by connecting smaller ‘pods’ consisting of several directly connected servers, provides the ideal underlying infrastructure to start to develop converged networking. Audit your architecture and validate that you can meet the latency requirements of your applications - if you cannot meet them now, you certainly will not be able to meet them in the future. Visibility and control
The goal of a converged network is to bring automation to the data centre by providing a coordinated approach Data Centre Hub March 2015 | 37
converged | Infrastructure
an entire corporate IT infrastructure ensuring it can seamlessly support real-time applications - and this will be essential to businesses of any size in order to deliver mission-critical applications securely to employees, no matter where they are. Vendor agnostic Thirdly, vendor agnosticism is the key to customisation and programmability - make sure you aren’t locked in! One of the fundamentals of true SDN is that it should be vendor agnostic, allowing for control and management of the network through SDN controllers, rather than through separate vendor silos that require administrating individually using proprietary protocols. The promise of SDN is to create the ability for customisation and programmability that can deliver the functions, applications and services that are unique to each organisation. WThis is why it is vital to use hardware from vendors that use standards based technology, so as to avoid the classic vendor lock-in and be restricted to proprietary technology from one vendor, or a limited handful of select third party partners. This is why open protocols such as OpenFlow - enabling SDN through the management of network hardware from multiple vendors - or OpenStack’s software tools for building cloud computing platforms are so important, as they provide the freedom to build and 38 | Data Centre Hub March 2015
customise networks as businesses see fit. With closed or propriety protocols you will always have restrictions on what customisations you can make, and the levels of programmability you can achieve, hindering the flexibility that SDN is designed to offer networks and service delivery. Enpowerment Finally, empowering the user is the goal of network services - but will you be able to? One of the most important to things remember is that a network is more than just a data centre, and it goes from the core all the way to the end user, wherever they are. A business needs to have the required capacity across the whole network. If modern business networks are about application and service delivery, then the user experience must feature highly in the minds of providers and solution architects. Converged networks need to focus on the entire user experience. A holistic view needs to take into account the access infrastructure from data centre to WLAN, LAN and cabling to WiFi infrastructure in office locations. The key question to be asked is whether the converged network will have the programmability capabilities for end-to-end delivery of applications - and with the right priorities for a corporate environment. Here a Unified Access (UA) solution is essential, supported by a converged Application Fluent
Network to deliver secure, high quality and consistent applications. UA provides a single policy infrastructure that does everything from automating the on boarding of new employees with the correct levels of access and device control, ensuring QoS across the entire converged network. Unifying the management and user experience across wired and wireless environments brings visibility and control of the applications across the whole network, supporting the prioritisation of business critical functions which ensures employee have the high quality tools they need to perform. Meet your potential It’s clear that network infrastructures need to adapt if they are to keep up with rapid developments we are seeing with multiple devices and multimedia applications being brought into the workplace. Converged networks certainly provide the answer, but organisations must consider the four key questions above before opting for a network upgrade - not all solutions are made equal, and it pays to set off on the right foot, from the word go. www.enterprise.alcatel-lucent.com
DATA CENTRE SUMMIT 2015 NORTH
Data Centre Summit North is the first in a series of new one-day conference focussed events, with the first set to take place at Manchester’s Old Trafford Conference Centre on the 30th of September 2015. DCS will bring all of the industries thought leaders together in one place, with the industries leading vendors and end users. The focus of the event will be placed on education, networking and debate, It will provide an open forum for delegates to learn from the best in the business. The event will also feature an exhibit hall where the industries leading companies will all show their newest products and services, together with a networking lounge so that you can make connections with like minded business professionals. Schneider Electric White paper
To enquire about exhibiting call Peter Herbert on 07899 981123 or Ian Titchener on 01353 865403
om .c ld or
Manchester’s Old Trafford Conference Centre
ew tr en ac at .d
w w
w
30th of September 2015
DATA CENTRE SUMMIT 2015 NORTH Platinum Headline Sponsor