Tag: Ascent initiative

Returns From Lincoln Ascent

Some of the numbers have started to come in from work done for Lincoln Ascent and things are looking promising to this point. The parts that are rounding up right now focus on speeding up Moab’s scheduling iteration and reducing the network overhead between TORQUE and Moab. One of the excesses in network communication between […]

Announcing the Beta Release of Moab 8.1

This entry is part of 7 in the series Adaptive Computing at SC14

This entry is part of 7 in the series Adaptive Computing at SC14We’re proud to announce the beta release of Moab HPC Suite – Enterprise Edition (Moab 8.1) at the SC14 conference in New Orleans. The Adaptive Computing team is providing sneak peek demos of Moab 8.1, Viewpoint admin portal, elastic computing functionality and advancements […]

Game Changer for HPC Scale and Performance

The exascale wave in today’s HPC market is creating an inflection point, where familiar solutions are simply inadequate. Modern supercomputers have so many internal network interconnects and coordinate so many calculations at such a rate that traditional scheduling cannot keep up. Jobs sit idle when they should be running; policy constraints are lost in the […]

Lincoln Ascent: Communicating Between Moab and TORQUE

One of the major development projects for TORQUE 5.0 and Moab 8.0 was the Ascent project. We were extremely pleased with the progress made during this effort, so we’re continuing the effort into the next release (codename: Lincoln), and I’d like to take this chance to introduce it. The purpose of Ascent is to rapidly […]

Kilby’s Net Effect on TORQUE

Overall, the Kilby release of TORQUE (TORQUE 5.0) should provide a faster, more stable TORQUE with some important new features. TORQUE should be more stable as we’ve expanded the breadth and depth of tests; it is significantly faster as a result of the work done for the Ascent Initiative; finally, several new features have been […]

Tuning The Engine: Moab Speed Improvements

Did you ever ride one of those single-piston two-stroke motorcycles? We used to call ours “Thumper” —it had a lot of torque and did a great job getting you up a hill, but wasn’t necessarily the fastest from zero to 60MPH and wasn’t very pleasant on the highway. Modern sport motorcycles have four or six […]

An Update on Ascent for TORQUE

Since my last blog post on the Ascent Initiative there have been a lot of new improvements and we have a lot more data. The following chart provides data on two improvements previously mentioned but unquantified as of my previous post. This relates to having the jobs report mom status instead of the server polling […]

The Dawn of Parallel Moab

One of the great ironies of the HPC industry is that the scheduling technologies that keep massively parallel supercomputers busy are, themselves, mostly serial. Moab and its competitors have their roots in theoretical work first productized in the 1980s and 1990. At that time, computer science mainly used parallelism for enormous matrix math problems—not for […]

Put Some Nitro In Your HPC Engine — Announcing MTM “Early Availability” Preview

All of us need to get more work done, faster. This is certainly true in traditional HPC, and it’s also imperative in map/reduce clusters, render clusters, and clouds. The dawn of exascale computing and the juggernaut of Big Data both demand that we get comfortable with millions of jobs in a queue, and thousands of […]

The Ascent Initiative: Increasing Throughput and Scalability

The ascent initiative is designed to make TORQUE and Moab increase in throughput and scalability across the board. Our goal is that we’ll increase both by 2-3 times across the board in the coming release; we expect to meet this goal, and we expect that for certain use cases we will be able to speed […]