This article is more than 1 year old

Cluster-grappling kids clash: Battle of the Big Iron in the Big Easy

Bring your HPC weapons to the Mississippi

Kraut Klusters

Team Kraut (FAU, Germany) isn’t politically correct, but it's a lot of fun, and it knows its HPC too. It picked the name “Team Kraut” itself and is using it on its competition paperwork. This fact alone makes them at least 27 per cent funnier than most other Germans.

This is the second time Team Kraut has made a run at the SC cluster cup. It learned a lot from its 2013 outing and it shows in its cluster and approach to the competition.

Last year, the team had a behemoth cluster with a raft of cores and heaps of GPUs. While it had a respectable finish, it realised that its everything-plus-the-kitchen-sink approach might not be the winning strategy.

This year, the team showed up with a sophisticated six-node design that mixes four regular compute nodes (or Kompute Nodestats) with two uber compute nodes (or Uber Kompute Nodestats).

The regular compute nodes feature dual E5 14 core processors, running at 2.3GHz, with 128GB memory per node – straightforward stuff. The two uber nodes are where things get interesting.

Each uber node is equipped with dual E5 12 core processors running at a higher 2.8GHz frequency than the CPUs in the regular nodes. The two uber nodes are also decked out with four NVIDIA Tesla GPUs each, which is a lot of concentrated GPU goodness.

This configuration gives the team the ability to push GPU-centric apps onto the uber-nodes, while using the traditional regular nodes for standard non-CUDA-ised codes. When it's not running GPU-centric apps, it can idle the uber nodes and crank up the compute power on its CPU-only regular nodes.

It’s an interesting approach and has some merits – but, I think, is highly dependent on the application mix between GPU-centric and CPU-only workloads. If there’s a healthy mix of the two, Team Kraut will be able to run the apps in a way that maximises its cluster utilisation. But if the mix skews strongly in one direction or the other, then all bets are off.

Team USTC (University of Science and Technology – China) is fresh off an impressive début performance at ISC14 last June. As the new kids on the block in Leipzig, there wasn’t a lot expected of them. (Much like the also unheralded German 1990s boy band “Leipzig Boys on the Block". Its first single, “Teuton Minen Own Horn”, never made the charts. Although I think that Robbie Williams did cover it eventually. But we digress ...)

However, unlike the doomed L-Boys, USTC did much better than anticipated at its first student cluster competition. It finished just behind Overall Championship winner South Africa, and barely ahead of the much more experienced Team Tsinghua to grab a second place slot.

At the ISC competition in June, the USTCers were driving a heavy weight ten-node, eight K40 GPU behemoth. In a surprise move, it radically changed its system for SC14 – going with five nodes of Sugon’s latest liquid cooled kit, sporting 14-core Xeon E5 processors (140 total cores) clocked at 2.4GHz, with 640 GB RAM, and ten NVIDIA K40 GPUs.

This configuration potentially packs quite a punch, provided that the USTC kids are able to optimise the applications for use with GPUs.

Team Calderero (Purdue/EAFIT) consists of students from Purdue and EAFIT, a private university located in Medellin, Colombia. Purdue is no stranger to student cluster competitions, having participated in eight previous battles, but it’s an entirely new experience for their comrades in arms from Columbia.

Purdue certainly could have raised a cluster team of its own; it's done it eight times before. But this year, it decided to reach out to its sister school and pull them into the cluster competition fray.

By all accounts, it’s been a great partnership, with the Colombians bringing both enthusiasm and solid technical skills to the team. More importantly, EAFIT’s participation has spurred greater interest in HPC among university faculty and students.

Hardware wise, the Purdue/EAFIT team brought a modest six-node, 168 core system, with only 64 GB of memory per node. Although this sounds like it’s a little light when compared with other teams, La Boilermakers crammed six NVIDIA Tesla K40 GPUs into its cluster.

This is the first time any Purdue team has deployed GPUs in a competition system. I get the sense that this is more of an experiment this year, rather than a full-fledged competitive move.

More about

More about

More about

TIP US OFF

Send us news


Other stories you might like