HPC in the Cloud


Dedicated to covering high-end cloud computing
in science, industry and the datacenter

Language Flags

Credit Suisse Lineage Makes DynamicOps a Contender


The virtualization management market is hot, no one is denying that. But as with most things in life, great theories don't always pan out in practice; technologies can seem flawless in the lab only to fizzle when put to the test in real datacenters. However, this isn't necessarily the result of any ineptitude on the part of the development team. Rather, it is merely the result of the developers not having years of experience dealing with the hodgepodge and complexity that is the real world of enterprise computing.

This real-world experience is the very thing that could make Credit Suisse spin-off DynamicOps, with its Virtual Resource Manager (VRM) product, a virtualization management powerhouse. It's difficult to replicate in the lab the conditions and concerns of an international banking goliath.

The Catalyst: Virtualization

It all began in 2002, when Credit Suisse product development head Leslie Muller, now DynamicOps CTO, saw VMware explode onto the scene, signaling x86 virtualization's rise to prominence as the enabling technology for the next-generation datacenter. But it wasn't all roses for Muller, who “has a sense of déjà vu when that happened, thinking back to around the time when [Windows] NT 3.51 really made a splash in the market.” NT 3.51 was a great product, notes Muller, but the management piece was “a nightmare.” He saw the same fate awaiting virtualization.

The management concerns, however, weren't nearly enough to squelch Muller's optimism, as he foresaw virtualization being far too valuable to Credit Suisse to ignore it – especially in terms of time-to-market. In the old-guard way of doing things, businesses would develop new opportunities and want to execute within a couple of weeks, only to spend between three and six months waiting for a physical server to be requisitioned and ready to use. Alternatively, Muller saw businesses buying premium servers in bulk up front, only to find them obsolete by the time they needed to roll them out. Virtualization, on the other hand, would enable easy service delivery and truly flexible businesses. Like in the real world of ATMs and online shopping, all anyone wanted was self service, said Muller.

By the time Muller gave the entire cast of IT decision-makers at Credit Suisse a “shoot for the moon” speech about virtualization in 2005, his team already has been building prototypes and working, albeit quite unsuccessfully, with other virtualization and IT solutions providers to try and solve some of Credit Suisse's most pressing issues. Muller's team was laying out use cases and problems, asking companies like VMware, Microsoft and HP, among others, what technologies they had to address these issues. “To be honest,” he stated, “the answers we got were really unsatisfactory.” And even if they understood the technological aspects, they didn't understand the complexity involved in deploying virtualization across such a large company. So Credit Suisse took on the task of building VRM.

Having proved its value in short order, VRM received more internal funding and evolved into version 2.0 by mid-2007, at which time Credit Suisse took a look around the marketplace to see if there were any available solutions that could compete on the scale it needed. There were not.

Thus, partially as a result of clamoring from financial services competitors who knew of VRM and saw how it enabled Credit Suisse to out-compete them by moving quickly without having to buy and provision hardware, the company decided to sell VRM. The only way to do this correctly and ensure adequate resources for further development was to spin-off the company, seeking external funding and the whole nine yards. The result: DynamicOps -- Credit Suisse's first technology spin-off. (But, according to Muller, not its last.)

Operational Expenditure or Operational Experience?

Technology-wise, says Muller, VRM is designed to help with “cradle to grave” automation, deployment and management of virtual resources, including decommissioning old VMs. It sports an open architecture that works across leading hypervisors (VMware, Citrix, Microsoft, Sun) and operating systems, and can manage very large-scale, highly complex datacenters spanning into the hundreds of thousands of VMs (or merely hundreds). Its focus on facilitating self-service within organizations leads to drastically reduced operational expenditures, a natural result of the added complexity virtualization can bring with it, says Muller. Especially when you're dealing with geographically distributed companies with many different types of users, process automation is a bear. Different countries have different compliance regulations and different data retention policies apply to different types of users, explains Muller, all of which raises operating costs.

However, as good as this all sounds, the experts aren't too impressed with the capabilities of VRM -- they're impressed with the fact that they know it works as advertised. Everyone but VMware will soon support multiple virtualization platforms, says Rachel Chalmers, research director for The 451 Group's Infrastructure Computing for the Enterprise division. Additionally, Enigmatec, OpTier and Opalis all do run book automation, she says, and Embotics, Fortisphere and ManageIQ all do similar things with lifecycle management. Even VMware does similar things within its product family. “This is the most exciting, competitive landscape with managing virtual machines at the moment,” she says.

Jeff Byrne, senior analyst and consultant with the Taneja Group, names the same companies as worthy competitors, adding that Microsoft will be getting into the virtualization management game, as well. It is a “very competitive market,” he says, with each vendor having its own sweet spot. From a product perspective, VRM stands apart with its focus on simplicity and self-service, he says, but the real differentiator is VRM's conception within the walls of Credit Suisse, its real operational experience.

Having spent seven years at Credit Suisse and even more before that, Muller knows well the process of evaluating products and vendors, and now that the shoe is on the other foot, he sees Credit Suisse's street cred paying dividends. “Sitting on the other side of the fence when I come to these companies, I see them following a slightly different trend,” he says. “These companies understand very quickly that we actually have that operational experience, and we have the experience of doing virtualization on a massive scale at Credit Suisse.” DynamicOps is entirely a standalone company, noted Muller, but “we are shameless about utilizing our relationships through Credit Suisse.”

What the Future Holds

From a product perspective, the No. 1 trend is increased scale. Early adopters of virtualization have gone aggressively after the low-hanging fruit of server virtualization, consolidating legacy systems and batch processes, says Muller, but many of these companies are becoming stuck as they attempt to virtualize all of their servers. The barriers are both cultural and technological, he says, and a lot of work still needs to go into hypervisor technology to make this happen.

But the discussion of scale does not stop at the server. Muller says customers are talking about virtualizing tens of thousands of desktops, a process that tends to be more complex than server virtualization due to various users, types of users and use cases. For the sake of reference in terms of scale, Muller says Credit Suisse itself has some 20,000 desktops, and some potential customers are are looking at hundreds of thousands of desktops with an eye toward virtualization.

Beyond that, says Muller, DynamicOps will be looking at how to bring together grid technology and virtualization. There is a huge amount of growth that has to take place here, he believes, because grid computing is very important to business, and can benefit from the capabilities of virtualization. Grid computing has made a huge difference to Credit Suisse in terms of fast reaction time and revenue generation, explains Muller, but businesses cannot keep infinitely growing their grids because most of these companies also are looking to reduce power consumption -- often through virtualization. Muller believes he and his cohorts at DynamicOps can leverage their experiences in managing large, heterogeneous, service-oriented grids at Credit Suisse to help bridge this divergence.

Chalmers, who is familiar with Credit Suisse's grid practice from her days with The 451 Group's Grid Adoption Research Service, says it's not surprising that DynamicOps has this grid-inspired vision. “It totally makes sense that the people who already have built and deployed these grids would be the right people to turn to for exactly the same kind of automation and resource management,” she said. “Except instead of doing very CPU-intensive workloads across a few machines, they’re doing virtual workloads across any machine within the enterprise.”

In addition, says Chalmers, VRM's transparent nature will allow customers to take their grid and virtualization initiatives and evolve them into the IT trend du jour: the cloud. IT shops have “service provider envy,” she explains, and they want to deliver infrastructure like Rackspace or Amazon, but you absolutely need transparency to do this. “It's not until you know what you're delivering that you can start to do costing and chargeback and billing.”

Muller mentions similar objectives, stating as the “end goal for VRM” the realization of the datacenter 2.0 vision, where everything is is virtualized and dynamic. DynamicOps is working with networking, storage and application streaming vendors to make this happen, he added.

Of course, none of these technological advances can come to fruition if DynamicOps doesn't grow the customer base to keep itself in business -- but that doesn't seem to be an issue. In fact, the only real question seems to be whether the company will be able to overcome its financial services roots and break into the broader marketplace. (Not that the financial market isn't a lucrative niche in which to be stuck.)

Taneja Group's Byrne doesn't foresee it being a challenge, stating that DynamicOps has all the ingredients necessary to be a leader in the space. “I would much prefer to see a company start with a market like the one they’re starting with and be able to leverage that rather than starting with, let’s say, a state government market or retail,” he said. “I think this gives them a lot more ‘hardening’ in the early stages than they would get otherwise."

Ask Muller and he'll make it clear that financial service is “absolutely not” the target market for VRM, but it is the first and current target due to DynamicOps' contacts from Credit Suisse. Other targets are in the medical, media and manufacturing industries, as well as anyone doing virtualization at scale, he explained. In fact, in the five months since its incorporation, DynamicOps has picked up more than half a dozen proof-of-concepts, with more scheduled within the next few weeks, and has already installed VRM at several customer sites. The customers have been a mixture of tier 1 banks, as well as medical and mid-market companies, said Muller.

It doesn't hurt that DynamicOps was born with a silver spoon in its mouth, says Chalmers, and because everyone knows who they are, “the game is theirs to lose.” “If they don't acquire 10 customers in the next six months,” she predicts, “they're doing it wrong.”

Most Read Blogs

Aspen

Short Takes

NASA Builds 'Climate in a Box'

May 23, 2013 | The study of climate change is one of those scientific problems where it is almost essential to model the entire Earth to attain accurate results and make worthwhile predictions. In an attempt to make climate science more accessible to smaller research facilities, NASA introduced what they call ‘Climate in a Box,’ a system they note acts as a desktop supercomputer.
Read more...

Running Computational Fluid Dynamics in the Cloud

May 16, 2013 | When it comes to cloud, long distances mean unacceptably high latencies. Researchers from the University of Bonn in Germany examined those latency issues of doing CFD modeling in the cloud by utilizing a common CFD and its utilization in HPC instance types including both CPU and GPU cores of Amazon EC2.
Read more...

In Support of Cloud-based Rendering

May 10, 2013 | Australian visual effects company, Animal Logic, is considering a move to the public cloud.
Read more...

Internet2 Awards Program Seeks Innovative Applications

May 10, 2013 | Program provides cash awards up to $10,000 for the best open-source end-user applications deployed on 100G network.
Read more...

Sponsored Whitepapers

Best Practices in Big Data Storage

05/10/2013 | Cleversafe, Cray, DDN, NetApp, & Panasas | From Wall Street to Hollywood, drug discovery to homeland security, companies and organizations of all sizes and stripes are coming face to face with the challenges – and opportunities – afforded by Big Data. Before anyone can utilize these extraordinary data repositories, however, they must first harness and manage their data stores, and do so utilizing technologies that underscore affordability, security, and scalability.

Exploring the Potential of Heterogeneous Computing

04/02/2012 | AMD | Developers today are just beginning to explore the potential of heterogeneous computing, but the potential for this new paradigm is huge. This brief article reviews how the technology might impact a range of application development areas, including client experiences and cloud-based data management. As platforms like OpenCL continue to evolve, the benefits of heterogeneous computing will become even more accessible. Use this quick article to jump-start your own thinking on heterogeneous computing.

Sponsored Multimedias

Newsletters

Stay informed! Subscribe to HPC in the Cloud email Newsletters.

HPC in the Cloud Update
HPCwire Weekly Update
Digital Manufacturing Report
Datanami
HPCwire Conferences & Events
Job Bank
HPCwire Product Showcases


ISC

HPC Job Bank


Featured Events



  • June 16, 2013 - June 20, 2013
    ISC'13
    Leipzig,
    Germany

  • June 17, 2013 - June 18, 2013
    Forecast 2013
    San Francisco, CA
    United States




HPC in the Cloud Conferences & Events