[Most Recent Entries] [Calendar View]
Friday, August 25th, 2017
- “To ensure that the availability of the data center remains constant during this time, we make sure we have fuel. Fuel should always be checked. We make sure we have certain spare parts and tools that we need or anticipate that we may need. We make sure that everything is anchored down — if the wind comes in, you don’t want things flying around. Is there food and water? Is there bedding, if the staff needs to stay on-site, and communications equipment?
- “Also, if something really bad is coming in, we will establish an emergency control room, where we will actually co-ordinate all events from. And all staff are fully aware of their roles and responsibilities in relation to the control room.
- “When an event hits, then we constantly monitor. It’s a constant monitoring of the systems to make sure the availability and critical loads are at the correct levels. And we make sure that we have the right personal protection equipment as well. Sometimes they need to wear it and have it with them, on their body, because they never know when they’re actually going to be needing it in the fields during an event like this. And again, monitor the news consistently. Monitor the weather channels.
- “Something else that’s important is a buddy system. Do not let teams go work alone. Be with them, in case something goes wrong and you need additional support.”
| Time | Event |
| 12:00p | A Liquid Cooling Renaissance Coming to the Data Center? With 35-plus years as a full-service mechanical and mission critical environments contractor with a heavy emphasis on service, maintenance, and repair, I specialize in mission critical cooling (heat rejection) and electrical infrastructures, as well as the comfort cooling surrounding them. As you know, seemingly everything in the world is becoming digitized. A talking appliance that sits on your counter top and does just about anything you ask “her” to do: play music, order pizza, or simply sing to the baby. If you can think of it, it can, has, and will become some form of synthetic and augmented reality. Welcome to the world of “Big Data.” As part of the heat rejection/cooling industry, we cannot sit idly, watching the world go round; and we aren’t. New, increasingly more powerful technologies dig ever deeper and produce a lot of data — data that must be housed, analyzed, delivered, and stored in some form of a data center. It could be a colocation facility, your on-premises data center, an edge computing node, or a public cloud. Whatever the deployment, we must continue to work hard to understand today’s data center topologies. Air-cooled, water-cooled, or liquid cooled, it is all about heat rejection. See also: Deep Learning Driving Up Data Center Power Density Watching the data center cooling market over the last several years, there is currently a push toward liquid cooling that hasn’t been seen before. Why? Because IT rack power densities are growing. Five years ago, 5kw worth of load was a pretty well-loaded rack; today, that’s mid- to low-density. It is difficult for many firms to achieve much north of 10kw with air cooling, mainly because most facilities were not designed such hot components. The natural solution? Use fluid. As we know, fluid is much more efficient in heat-transfer ratios versus air. Liquid cooling also comes in several forms. For example, it can be used in a rear-door heat exchanger. There are also techniques and designs that mount a heat exchanger onto the CPU, GPU, or FPGA and extract the heat directly from the chip. Some manufacturers do it with Novec 7000, a non-conductive fluid; some manufacturers do it with regular water. These are just additional ways to reject heat without using air. There are also a few solutions that remove heat by submerging servers in a tank of fluid. Within the last few weeks I have had conversations with three liquid cooling-based companies that have a pipeline full of opportunities. This particular sector of the industry is to better utilize data center space, both in terms of square footage and within each rack. The thought leaders behind this newest wave in testing are in fact finding they can rack and stack at much higher densities than before. Next up, cryptocurrency perhaps? This too is an ever-increasing portion of the IT community. It will be interesting to see how these players fare down the road. Will the frequent mining hardware refreshes continue? Will we see that sector move to a more liquid-centric mindset? Liquid-based heat rejection is a mystical to some and appealing to many; but there’s still a lot of “aquaphobia” that must be dealt with. Over the years this phobia has grown and waned, and depending on your particular age and/or time in the IT trenches, you may have observed that dynamic. As a 55-year-old, I saw mainframes sitting around in the data center in the early years. Most of these, especially the big IBM mainframes, were chock-full of chilled water. With the onset of the PC/desktop generation, water and any liquid systems started being perceived as a big n-no. But technology is moving forward, and the younger generation and not overly seasoned IT crowd see the virtues of liquid. Regardless of your biases, either pro or con on liquid, it is coming in an impactful way. I, for one, am excited to see where we go from here. Until next time … AFCOM member Greg Crumpton has 35-plus years as a full service mechanical and mission critical environments contractor with a heavy emphasis on service, maintenance and repair. He specializes in mission critical cooling (heat rejection) and electrical infrastructures, as well as the comfort cooling surrounding them. |
| 5:00p | Harvey: Hurricane Preparation Tips for Data Center Managers As Hurricane Harvey bears down on the Texas coast, expected to make landfall around Corpus Christi either tonight or Saturday morning as a dangerous Category 3 storm, the men and women who work in data centers in the area are undoubtedly earning overtime as they prepare for the storm’s onslaught. Keeping data centers operational during natural disasters can be critical to the health and safety of the affected area’s residents, as they supply the lines of communications for many first responders and provide access to valuable information about weather conditions and the state of the area’s infrastructure. During pending disasters such as this, employees from Schneider Electric’s various data center divisions can often be found on the scene, offering their expertise to help data centers successfully get through the emergency. They’re good to have around, because as the old saying goes, they’ve been there and done that — countless times. A month or so ago, Data Center Knowledge talked with David Gentry, Schneider’s VP of data center services, and Mark Rentzke, senior manager for its global data center services for Europe, the Middle East and Africa, on the subject of hurricane preparedness for data centers. Both Gentry and Rentzke stressed that getting through an event such as a major hurricane unscathed depends on planning that begins well before the emergency arises. “You have to make sure that all staff are trained on emergency preparedness for events that you are anticipating or not anticipating,” Rentzke said. “Make sure the staff understands the contingency plans, the escalation plans, the coordination from security. Make sure that if you need to load-shift any potential equipment, you know which equipment needs to be load-shifted, if you need to drop power at some point, or lower the power.” “Once an event takes place, it’s obviously too late to acquire needed supplies,” Gentry added. “It’s also too late to assure that all the required equipment is available in top working condition — things that you might need to get through a big storm. You need to do some things upfront to assure that everything is fully charged and to train personnel on things such as proper communications and equipment operations. During an event like this, which is high-anxiety, it’s better if you can already have seen through and [given] those people on the site some points at which they know that they have to do a certain thing.” Rentzke supplied us with a laundry list of things data center operators need to be prepared to handle: One of the things we wondered about was preparation for the flooding that can be common during hurricanes. This might be particularly pertinent in the case of Harvey, which is expected to stall once it makes landfall and drop more than 35 inches of rain on the region. “If a rack is flooded, none of the servers are going to be operating because you’re going to have some very significant electrical problems,” Gentry said. “But you should make sure the most critical servers are grouped together and are on a dedicated source, so that you can isolate them. Isolating those most important servers would definitely be an important thing to do so that you can maintain power, cooling, and whatever is necessary to those servers and that area of the data center for as long as possible.” “A data center that is established and with a mature program in place will have a risk register,” Rentzke added. “Within the risk register they will identify any major risks. For example, if they have a low area within the campus, they might make note of that, and mitigation for that may be putting pumps in place. Preparedness for this would be to make sure those pumps are running.” Rentzke pointed out that even if a data center survives a catastrophic storm seemingly unscathed, there still might be hidden equipment damage, caused perhaps by an electrical surge that went undetected during the storm. “After the hurricane, we need to look at all the equipment,” he said. “In some cases, depending on the critical nature of the hurricane that came through, we need to make sure that the equipment is running at its optimum performance. We might even need to have certain equipment checked out and have a little bit of a deep-dive into the actual health of the equipment. Sometimes after a hurricane you can find stuff lurking within systems that you weren’t aware of that could be danger points later down the road. So there’s an increased risk of something going wrong after the storm as well.” Ultimately, according to Rentzke, experiencing a major hurricane or any other catastrophic event turns into a learning experience for the data center operators and their emergency-preparedness teams to better prepare the facility for the next major event. “The whole thing revolves around a continuous improvement, a lessons-learned program,” he explained. “If anything is identified as a corrective action that we can improve on for next time, we put that into our lessons-learned program and make sure there’s a continuous improvement in our methodology for operating a data center in hurricane conditions.” As the storm approaches, however, the major concern is about well-being of the men and women who will potentially be putting themselves in harm’s way during the storm. “The safety of our staff, the well-being of all the personnel on site, is our ultimate goal,” Gentry said. |
| 5:17p | Artificial Intelligence and the Future of HPC Simulations Robert Wisniewski is Chief Software Architect, Exascale Computing at Intel Corporation When discussing artificial intelligence and how it relates to the future of high-performance computing (HPC), it’s important to begin by noting that while machine learning and deep learning are sometimes viewed as synonymous, there are distinguishing features of deep learning that have allowed it to come to the forefront. Machine learning can be viewed as a sub-field of artificial intelligence, with a core underlying principle that computers can access data, recognize patterns and essentially ‘learn’ for themselves. Deep learning takes this one step further by increasing the depth of the neural network and providing massive amounts of data. The combination of the increased computational power of modern computers and the recent deluge of data from a myriad of sources has allowed the deep learning sub-discipline to produce impressive results in terms of capability and accuracy. Within the HPC community we have witnessed the uptake of both machine learning and, in particular, deep learning. Given the successes mentioned above, there is a high likelihood in the near future that we will see an increasing number of HPC computation cycles being used for machine learning, deep learning, and any other artificial intelligence-type of capability. For people designing the computers and the system software to support them, it is important to realize that while AI is revolutionizing computation and is exciting, a large percentage of HPC cycles will continue to be required by traditional simulation – and even the end state is likely to be a combination of AI and simulation as they are symbiotic. Thus, the move toward AI needs to be approached in a balanced way because while there is no doubt that it will form an important and increasing portion of HPC, it is not going to replace classical simulation cycles. Simulation cycles will remain, and on large HPC machines they will continue to be a critical step. The difference is that there will now be an additional class of cycles that includes machine learning and deep learning. Ten years ago, the amount of data being generated was only just beginning to get to the point where machine learning and deep learning algorithms could be used successfully. Now that there is a critical mass of data and the ability to store it and access it efficiently, significant strides can be made in many scientific fields. For example, in climatology, scientists have been able to couple the massive amounts of data with deep learning algorithms to discover new weather phenomenon. This is just one discipline in which huge amounts of data is generated by simulations, and the coupling of deep learning algorithms is starting to form an important tool base for scientific discovery. Trends like this indicate that machine learning and deep learning algorithms will increasingly be run in conjunction with existing HPC algorithms, and will help them analyze the data. By combining the capabilities of a cohesive and comprehensive software stack that runs and manages HPC systems from small turn-key machines to large supercomputers with those of machine-learned algorithms, enterprises can benefit from significant computational savings in terms of both time and effort. Redundant system administrator tasks can be eliminated, software upgrades facilitated, and time can instead be devoted to customization and the programming of machine learning algorithms relevant to an organization’s needs. Intel HPC Orchestrator democratizes HPC and aids entry into the machine and deep learning space by easing the burden of building and maintaining HPC systems. It allows organizations the opportunity to leverage HPC capability, and provides a more efficient ecosystem by eliminating duplicative work across organizations and by allowing hardware innovations to be realized more quickly. The efficient enabling occurs because Intel develops the software stack in parallel with the hardware. In addition to providing the composability of frameworks across HPC and AI, we are investigating technologies to more tightly couple the application in an HPC AI workflow. Depending on the granularity of the couple, different technologies are needed. For example, if users are needing to loosely couple their applications across a machine, the resource management infrastructure may need to be enhanced. For more complex scenarios, such as running a simulation algorithm concurrently with an AI algorithm with the requirement that they can access the same data without the need to spend time moving that data in and out of the system, then users will want tighter coupling; maybe even at a node granularity. Some researchers are already working on this with positive results. To make this possible, we need to have capabilities that allow us to take a given node, or a local set of nodes, and be able to divide it up in a manner that allows both the new artificial intelligence algorithm and the classical HPC simulation to run. Because the simulation portion is sensitive to noise (execution time variance causes by system software interruptions, etc. that induce a difference application progress), we need technologies that can isolate the simulation execution from the AI execution. While containers provide some amount of isolation, through our mOS architecture (an extreme-scale operating systems effort), we are looking to provide greater degrees of isolation. We are looking to take technologies such as mOS and make them available through Intel HPC Orchestrator, as well as contribute them to OpenHPC (www.openhpc.community). In addition to coupling HPC and AI applications, another area where this tighter coupling can be beneficial is uncertainty quantification, which allows scientist to provide tighter error bounds on simulation results. As an analogy from the previous weather topic, a prediction for hurricane path is not a single line, but rather a probably of likely paths. The reason is because the conditions affecting the hurricane’s path each have a certain error associated with them. When the sum of these errors are mathematically combined, the result is a hurricane plot that is commonly shown on weather maps. The ability to accurately represent the error on a simulation plays an important role in scientific studies. Uncertainty quantification involves running a set of lower fidelity simulations with different error condition models, and with varying initial starting points. Today, the couple between these runs are relatively course grain. In some fields, if the coupling can be finer grain, potentially even on an iteration granularity, much tighter and higher fidelity error bars are achievable. This is just one example of where coupling could benefit classical simulation. For machine and deep learning, there are many additional possibilities where tighter coupling can be beneficial. It seems clear that the path between machine and deep learning, and classical HPC simulation is converging. However, for the near future at least, machine-learned algorithms will remain best suited to filling in gaps of information, especially in areas where simulated interactions rather than predictions are needed. In the current computing landscape, algorithms that can learn and make predictions from incredibly large datasets are entering existing workflows and enabling enterprises to save costs, streamline processes, and drive forward innovation. Opinions expressed in the article above do not necessarily reflect the opinions of Data Center Knowledge and Informa. Industry Perspectives is a content channel at Data Center Knowledge highlighting thought leadership in the data center arena. See our guidelines and submission process for information on participating. View previously published Industry Perspectives in our Knowledge Library. |
| 5:48p | Multi-Cloud Is a Reality, Not a Strategy – Part 1 James Kelly is the Lead Cloud and SDN Expert at Juniper Networks. So you’re doing cloud, and there is no sign of slowing down. Maybe your IT strategies are measured, maybe you’re following the wisdom of the crowd, maybe you’re under the gun, you’re impetuous or you’re oblivious. Maybe all of the above apply. In any case, like all businesses, you’ve realized that cloud is the vehicle for your newly dubbed software-defined enterprise: a definition carrying onerous, what I call, ‘daft pressures’ for harder, better, faster, stronger IT. You may as well be solving the climate-change crisis because to have a fighting chance today, it feels like you have to do everything all at once. Enduring such exploding forces and pressures, many enterprises simultaneously devour cloud, left, right and center; the name for said zeitgeist: multi-cloud. If you’ve not kept abreast, overwhelming evidence – like years of the State of the Cloud Report and other analyst research – show that multi-cloud is the present direction for most enterprises. But the name ‘multi-cloud’ would suggest that using multiple clouds is all there is to it. Thus, in the time it takes to read this article, any technocrat can have their own multi-cloud. It sounds too easy, and as they say, “when life looks like Easy Street, danger is at the door.” Apparition du JourWhile some cloud pundits forecasting the future of IT infrastructure will shift from hybrid cloud to multi-cloud or go round and round in sanctimonious circles, those who are perceptive have realized that multi-cloud pretense is merely a catch-all moniker, rather than an IT pattern to model. As with many terms washed over too much, “multi-cloud” has quickly washed out. It’s delusional as a strategy, and isn’t particularly useful, other than as le mot du jour to attract attention. How’s that for hypocrisy? Just like you wouldn’t last on only one food, it’s obvious that you choose multiple options at the cloud buffet. How you choose and how you consume is what will make the difference to if you perish, survive or thrive. Choosing and consuming wisely and with discipline is a strategy. Is the Strategy Hybrid Cloud?Whether you want to renew your vows to hybrid cloud or divorce that term, somewhere along the way, it got confused with bimodal, hybrid IT and ops. Nonetheless, removing the public-plus-private qualification, there was real value in the model to unite multiple clouds for one purpose. It’s precisely that aspect that would also add civility to the barbarism of blatantly pursuing multi-cloud. However, precipitating hybrid clouds for a single application with, say, the firecracker whiz-bang of the lofty cloud bursting use case, is really a distraction. The greater goal of using hybrid or multiple clouds should be as a unified, elastic IT infrastructure platform, exploiting the best of many environments. Opportunities for OpsWhether public or private, true clouds (which are far from every data center—even those well virtualized) ubiquitously offer elastic, API-controllable and observable infrastructure atop which devops can flourish to enable the speeds and feeds of your business. An obvious opportunity and challenge for IT operations professionals is in building true cloud infrastructure, but if that’s not in the cards for your enterprise, there is still even greater opportunity in managing and executing the strategy for the wisdom and discipline in choosing and consuming cloud. In the new generation of IT, ops discipline doesn’t mean being the ‘no’ person, it means shaping the path for developers, with developers. This is where devops teaches us that new fun exists to engineer, integrate and operate things. It’s just that they are stacks, pipelines and services, not servers. Of course, to ride this elevator up, traditional infrastructure and operations pros need to elevate their skills, practicing devops kung-fu across the multi-cloud and possibly applying it in building private clouds too. Imagine what would happen if we exalted our trite multi-cloud environment with strategies and tactics to master it? We can indeed extract the most value from each cloud of choice, avoid cloud lock-in, and ensure evolution, but you probably wouldn’t be surprised to learn that there are also shortcuts with new technical debt spirals lurking ahead. To secure the multi-cloud as a platform serving us, and not the other way around, some care, commitment, time and work is needed toward forming the right skills and habits as well as using or building tools, services and middleware. There is indeed a maturing shared map leading to portability, efficiency, situational awareness and orchestration. In my next article, we’ll examine some of the journey and important strategies to use the multi-cloud for speed with sustainability. Stay tuned. Opinions expressed in the article above do not necessarily reflect the opinions of Data Center Knowledge and Informa. Industry Perspectives is a content channel at Data Center Knowledge highlighting thought leadership in the data center arena. See our guidelines and submission process for information on participating. View previously published Industry Perspectives in our Knowledge Library. |
| 9:01p | Digital Realty to Build Mega-Data Center in RagingWire’s Texas Backyard *Updated with details of tax incentives offered to Digital Digital Realty Trust is planning a gigantic data center campus in Garland, Texas. This will be the second data center build in town, which is part of the Dallas-Fort Worth Metroplex but hasn’t seen the kind of rapid data center market growth its neighboring Richardson, Plano, Carrollton, and Dallas have. The first data center in Garland was built by NTT communications-owned RagingWire, one of Digital Realty’s direct competitors. Garland officials have been hard at work trying to attract data center development to the city for years, and landing San Francisco-based Digital Realty — one of the world’s largest data center landlords, which lists Facebook, Equinix, IBM, AT&T, and Verizon as some of its biggest customers — is the most recent fruit of that labor. The city’s economic development department announced the project Friday. Digital is planning to build the 150MW campus in three phases, expected to cost $350 million each, according to Garland officials. It will be built on a 47.5-acre plot at 1702 West Campbell Road, literally across the street from the RagingWire campus, where the first data center was launched this past April.
Map courtesy of Garland Economic Development While players in the Dallas data center market spent 2016 watching hyper-scale cloud deals go to other regions, the first half of 2017 brought a wave of cloud companies taking data center space in DFW, resulting in a 50-percent increase in inventory absorption from the prior year, according to a recent North American market report by the real estate firm Jones Lang LaSalle. With 27MW of capacity leased in the market, DFW was second only to Northern Virginia in absorption during that period. In addition to cloud providers, the Dallas data center market sees lots of demand from enterprise sectors, such as oil-and-gas and healthcare. See also: RagingWire Takes Its Massive-Scale, Luxury-Amenities Data Center Model to Texas Garland’s pitch to data center developers is relatively cheap energy (the city owns and operates its own utility) and proximity to the Dallas data center market hotspots. It is just east of Richardson, south of Plano, and northeast of Dallas. RagingWire’s deal with the city also included property tax breaks and subsidies for fiber construction to link the campus to the data highways several miles away. David Gwin, Garland’s director of economic development, said in a statement:
Garland offered Digital Realty 40-percent tax breaks over seven years on business and personal property, and on real-estate valuation, Gwin told Data Center Knowledge in a phone interview. |
| << Previous Day |
2017/08/25 [Calendar] |
Next Day >> |