High-performance computing is the latest 'it' thing in the cloud
Every few months something new pops up as the latest and greatest feature touted by cloud providers. Earlier this year it was platform-as-a-service (PaaS) offerings. Recently, vendors have been pushing free or low-cost versions of their clouds. Now, perhaps it's time for high performance computing to have its turn in the spotlight.
This week, a new entrant into the market named ProfitBricks is coming forth with impressive network speed capabilities in its cloud, according to an independent review of the European company's performance. This falls in line with Amazon Web Services, which released a solid-state disk high input/output compute option over the summer.
ProfitBricks uses InfiniBand, a wired fabric that allows more than triple the data transfer rate between servers as compared to two of the industry's biggest players, Amazon and Rackspace, according to a test conducted by Cloud Spectator. "No one really comes up to their level in terms of networking speed in the U.S.," says Kenny Li, co-founder of the cloud analytics firm.
ProfitBricks is a European company founded in 2010 by Achim Weiss and Andreas Gauger, whose former managed hosting company, 1&1, was sold to United Internet and now is a leading international Web hosting company. It raised $18 million in venture funding, including from United Internet, and has had customers in beta in the U.S. and Europe until making its services generally available this week.
ProfitBricks allows users to customize their public cloud to their heart's content, providing a range of computing options that it says is wider than most of the big players in the market. Many cloud providers, including Amazon and Microsoft Azure, offer tiers of cloud computing resources that include a certain number of cores, virtual machines and memory. ProfitBricks instead allows users to choose as many cores and as much memory as they want, ranging from 1 to 48 cores and anywhere up to 192GB of memory. Some of Amazon's highest offerings, by comparison, include 88 Elastic Cloud Compute units, with 60GB of memory in its XL cluster compute offering.
On its back end, ProfitBricks uses Mellanox dual QDR 40Gb/s InfiniBand switches to connect the server and storage infrastructure, while using a KVM hypervisor and AMD Opteron processor. The company serves up 60 virtual machines per server, about a threefold increase over typical virtual machine per server deployments using 10G Ethernet connections, it says.
The company is one of the first cloud providers to use InfiniBand, Li says. While it's not new technology, analyst Bob Laliberte of Enterprise Strategy Group says InfiniBand just never really found a home in enterprise deployments or service providers. Some fast-trading Wall Street shops and research institutions may use it, but it's unnecessarily high performance for mainstream enterprises, he says. Ethernet's continued development has made that the default options for most deployments, he says, but a growing crop of HPC offerings could foreshadow InfiniBand's increased use.
Bob Rizika is heading up ProfitBricks's U.S. strategy and says the company is going after researchers, media/entertainment organizations with high graphic processing requirements, gaming sites and Web-based companies that need fast access to large databases. The company's intuitive graphic interface deployment chart aims to ease the management of new resources as well.
Melanie Posey, an IDC researcher, says ProfitBricks is carving out a niche for itself around HPC.
"Since ProfitBricks is coming into the IaaS up against established giants like Amazon and Rackspace (and now Google and Microsoft), they need to establish some differentiation in the market -- the technology is one of those differentiators," she says. Given the increased role analyzing large amounts of data could play in the future, she expects HPC offerings may continue to be an area that service providers look to up their offerings in. IBM, for example, she says is making a big push to provide data analytics software for use either on-site or as a cloud-based service.
There are a bevy of other providers offering similarly customizable solutions, including getting up into high performance workloads. CloudSigma, for example, is another European cloud provider looking to tap into the U.S. market with its customizable offerings. Having fully customized cloud offerings, like those available from ProfitBricks and CloudSigma, allows the providers to tailor solutions directly to the customer's need, says CloudSigma CEO Patrick Baillie. CloudSigma is looking to tap into the HPC trend itself, touting its work with European scientific organizations, including CERN, which over the summer led the work discovering the Higgs boson and used cloud resources from a network of providers, including CloudSigma to process the computing needs of the project through the Helix Nebula project.
Network World staff writer Brandon Butler covers cloud computing and social collaboration. He can be reached at BButler@nww.com and found on Twitter at @BButlerNWW.
Read more about data center in Network World's Data Center section.