Visit additional Tabor Communication Publications
January 03, 2013
InfiniBand had a banner year in 2012. It overtook Ethernet in the TOP500 list for the first time and established itself as the interconnect of choice in a handful of non-HPC applications. That was all good news for Mellanox, the dominant InfiniBand provider.
How good? Just take a look at the two charts in a recent GigaOM report that extolled the virtuous climb of InfiniBand over the past 12 months. The first chart illustrates InfiniBand switching places with InfiniBand as the number one interconnect for the world's top supercomputers. Actually, the chart is a little hard to read in that regard, but digging into the TOP500 numbers, one finds that InfiniBand now claims 224 systems, with Ethernet now significantly further back at 189 systems.
The second chart is even more telling. It shows Mellanox's InfiniBand revenue taking off in 2011 and then accelerating sharply in 2012. As the GigaOM piece implies, some of that revenue growth is being generated by Oracle's adoption of the technology in its popular Exadata product line.
Some of the accelerated growth in 2012 benefitted from a pent-up demand from customers who were waiting for Intel's Sandy Bridge Xeon CPUs to make their way servers.. That CPU came with built-in support for PCIe Gen3, which enabled compute clusters to incorporate the faster FDR (56 Gbps) InfiniBand gear.
The GigaOM report, however, posits that the bubble might just be a result of "delays in the availability of cost-effective 40Gigabit Ethernet and 100GigE," the implication being that once those solutions go mainstream, InfiniBand might lose some of its appeal. But as the author notes, InfiniBand still reigns supreme in latency, and for many HPC and distributed big data applications, that gives it a built-in advantage.
In any case, such heady growth cannot be maintained forever. And sure enough, the final quarter of 2012 was a bit of downer for Mellanox. On Wednesday, the company reported that its Q4 revenues are projected to be 119 million to $121 million, which is well below the company’s previous guidance of $145 million to $150 million. Investors hammered the stock, with the share price sinking nearly 14 percent on after-hours trading after the new guidance numbers were announced.
Despite that bit of bad news, Mellanox's annual revenue for 2012 will be around $500 million, which is nearly twice what is was in 2011. Mellanox blamed the bad quarter on "a weaker demand environment, challenging macroeconomic conditions, and a technical issue associated with FDR 56Gb/s InfiniBand cabling which caused approximately $20 million of FDR deployments to be delayed."
Mellanox is forecasting continued growth for this year, although not as steep as it enjoyed in 2012. Revenue in 2013 is expected to be well north of $600 million. The company may shed some additional light on what the future holds during its upcoming conference call on January 23 to discuss the fourth quarter financial results.
The Xeon Phi coprocessor might be the new kid on the high performance block, but out of all first-rate kickers of the Intel tires, the Texas Advanced Computing Center (TACC) got the first real jab with its new top ten Stampede system.We talk with the center's Karl Schultz about the challenges of programming for Phi--but more specifically, the optimization...
Although Horst Simon was named Deputy Director of Lawrence Berkeley National Laboratory, he maintains his strong ties to the scientific computing community as an editor of the TOP500 list and as an invited speaker at conferences.
Supercomputing veteran, Bo Ewald, has been neck-deep in bleeding edge system development since his twelve-year stint at Cray Research back in the mid-1980s, which was followed by his tenure at large organizations like SGI and startups, including Scale Eight Corporation and Linux Networx. He has put his weight behind quantum company....
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.
04/15/2013 | Bull | “50% of HPC users say their largest jobs scale to 120 cores or less.” How about yours? Are your codes ready to take advantage of today’s and tomorrow’s ultra-parallel HPC systems? Download this White Paper by Analysts Intersect360 Research to see what Bull and Intel’s Center for Excellence in Parallel Programming can do for your codes.
In this demonstration of SGI DMF ZeroWatt disk solution, Dr. Eng Lim Goh, SGI CTO, discusses a function of SGI DMF software to reduce costs and power consumption in an exascale (Big Data) storage datacenter.
The Cray CS300-AC cluster supercomputer offers energy efficient, air-cooled design based on modular, industry-standard platforms featuring the latest processor and network technologies and a wide range of datacenter cooling requirements.