December 17, 2007
“Cloud computing.” Merely utter the words and imaginations run wild. Minds fill with thoughts of computing fabrics rivaling the Internet in scale, desktop clients accessing a nearly limitless pool of resources located somewhere in the ether; and businesses experiencing a level of computing flexibility previously reserved only for these very same types of fantasies.
Well, it looks like those minds can just keep on dreaming, because when it comes to cloud computing, the future is not here today ... at least, not yet.
Although there has been endless discussion and speculation about cloud computing in the last few months – with a good majority caused by IBM announcing its Blue Cloud line of solutions -- it looks like the first solution to come to market under the “cloud” banner will localized in scale, hardly resembling the expansive Google-style infrastructure most associate with the IT world’s hottest new computing model.
Targeted for a spring 2008 release, IBM’s first Blue Cloud offering will be in the form of a BladeCenter featuring a combination of x86- and POWER-based blades, and which will be completely based on Linux and Xen virtualization and will take advantage of Tivoli provisioning management and monitoring technologies, as well as Apache Hadoop parallel processing software. “The objective with this first offering,” said Dennis Quan, IBM’s chief technical strategist for Blue Cloud, “is to give people a seed that they can use to start a cloud in their own datacenter.”
Of course, cloud computing, and Blue Cloud in particular, have thus far been presented as grand globally distributed solutions, so the unveiling of a single-site cloud solution might take some of the wind from the sails of believers -- and it might also serve to add substance to some analysts’ claims that cloud computing currently consists of a lot of smoke, but very little fire.
One such individual is Gartner vice president and distinguished analyst Thomas Bittman, who, at last month’s Gartner Data Center conference in Las Vegas, used the very same smoke and fire comparison to characterize the current state of cloud computing. And although he acknowledged the ultimate promise of cloud computing as a truly revolutionary architectural model, Bittman also impugned cloud solutions -- from Blue Cloud to past similar offerings like Sun N1/Terraspring -- on their lack of service level management and policy-based automation, commenting that for businesses which need maximum uptime and availability, cloud computing just doesn’t make sense right now.
David Mitchell, senior vice president of IT research at London-based analyst firm Ovum, agrees to a point. Early cloud computing offerings, he said, suffered from poor uptake because they lacked SLAs, and SLAs will again be important drivers for adoption of this next generation of cloud solutions. However, he noted, demand for solid and enforceable SLAs really will only be a deal-breaker in transaction-intensive and mission-critical environments, as in other application domains “a reduction in service quality short of a full outage is quite acceptable.”
Regardless of lax service-level requirements in many organizations, though, IBM is making progress on the SLA front. According to Mitchell, who has been following Blue Cloud since its inception within IBM’s High Performance on Demand Solutions (HiPODS) team (HiPODS actually implemented IBM’s internal cloud upon which Blue Cloud is based), IBM is addressing the issue by building more of the software management stack into the solution, thus allowing for better abilities in terms of service-level reporting, proactive management and self healing.
Said IBM’s Quan, who also heads up the HiPODS team, his group has been successful in making SLAs and policy-driven capabilities work on IBM’s internal cloud, and the company will introduce them into Blue Cloud solutions in due time. He added that IBM is working closely with some large financial services firms on Blue Cloud, which should say something about how much stock the company puts in SLAs and where it expects them ultimately to be.
Aside from financial services, IBM also is working with customers in the Web 2.0 and government sectors, although Quan believes cloud computing will have broad appeal across many industries. “Cloud computing and Blue Cloud are important because there is a convergence of a number of factors in the industry,” explained Quan. “Customers are running out of datacenter space [and] there are large increases in energy prices and management costs, and [that] is leading to the need [for] these large datacenters with highly refined scalable management characteristics and the ability to run these next-generation Internet applications efficiently.”
Mitchell has similar views about the market potential for offerings like Blue Cloud, noting that industries (e.g., financial services, Web 2.0, telecommunications, etc.) which experience high workload variability, high peak transaction workloads, high workload unpredictability and/or a need for rapid time to market will drive early adoption.
However, the fact that IBM is rolling out its Blue Cloud offerings on a smaller-than-expected scale certainly doesn’t mean global clouds (like that of IBM partner Google) aren’t on the horizon. In fact, said Mitchell, there is a natural evolution for this sort of thing. Much like what we have seen with grid computing, Mitchell’s evolution begins with an increased number of local facilities, which will evolve into regional offerings, which then will join nationally and/or globally. This also is analogous to the electricity market, explained Mitchell, where local offering grew into national networks, which then connected internationally. To follow this trend, he continued, cloud computing will need to focus on three things: (1) an increased focus on engineering standardization; (2) an increased focus on charging mechanisms for system usage; and (3) ensuring equitable and fair access to the infrastructure.
As for a time frame for all of this, Mitchell sees changes in the datacenter environment driving demand for cloud computing-type offerings for the next three years or so, with Web and Enterprise 2.0 companies driving adoption. And while there currently are a number of software vendors, large and small, vying to define the market, the Ovum view is that the definitions and associations given to cloud computing by large vendors like IBM, Sun Microsystems and HP are likely to rule the day.
So, what are we to take away from all of this? Are we to just lie down and accept the fact that the first generation of commercially accessible “cloud computing” offerings won’t allow anyone with the inclination and a little cash to scale like today’s Web giants? The short answer: Yes, but don’t let those grandiose dreams die just yet. After all, said Quan, small pockets of leading-edge users already have distributed resource fabrics. For the rest of us, though, “There is still a way to go with the bigger picture, the fabric vision,” said a down-to-Earth Quan, “and before we can get there, we have to get the building block pieces to come together.”
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.
May 10, 2013 |
Australian visual effects company, Animal Logic, is considering a move to the public cloud.
May 10, 2013 |
Program provides cash awards up to $10,000 for the best open-source end-user applications deployed on 100G network.
May 08, 2013 |
For engineers looking to leverage high-performance computing, the accessibility of a cloud-based approach is a powerful draw, but there are costs that may not be readily apparent.
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.
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.