TFLOPS PFS: Architecture and Design of a Highly Efficient Parallel File System

Garg, Sharad

doi:10.1109/sc.1998.10003

Cited by 6 publications

(5 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The architecture assumed is a high-performance parallel system similar to the Intel/Sandia Labs ASCI Red [25,10].…”

Section: System Architecturementioning

confidence: 99%

“…This study is motivated by earlier measurements of the parallel I/O performance on a prototype ASCI Red machine with up to nine I/O nodes and several hundred processor nodes at Intel [10,11]. These experiments showed that certain layouts of jobs relative to the I/O nodes saturated the communication bandwidth of some network links, resulting in serious parallel I/O performance degradation.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

The impact of spatial layout of jobs on I/O hotspots in mesh networks

Mache

Garg

2005

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

“…The architecture assumed is a high-performance parallel system similar to the Intel/Sandia Labs ASCI Red [25,10].…”

Section: System Architecturementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

The impact of spatial layout of jobs on I/O hotspots in mesh networks

Mache

Garg

2005

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

“…Indeed, processing speed has increased at a faster pace than parallel file system (PFS) bandwidth. For example, the first supercomputer in the Top500 list [1] of November 1999 was ASCI Red, with peak performance of 3.2 TFlops and 1 GB/s of PFS bandwidth [2]. Twenty years later, in November 2019, Summit had 200.8 PFlops (over 62, 000× faster) and a PFS capable of 2.5 TB/s (2, 500× faster) [3].…”

Section: Introductionmentioning

confidence: 99%

IO-Sets: Simple and Efficient Approaches for I/O Bandwidth Management

Boito,

Pallez,

Teylo

et al. 2023

IEEE Trans. Parallel Distrib. Syst.

View full text Add to dashboard Cite

One of the main performance issues faced by highperformance computing platforms is the congestion caused by concurrent I/O from applications. When this happens, the platform's overall performance and utilization are harmed. From the extensive work in this field, I/O scheduling is the essential solution to this problem. The main drawback of current techniques is the amount of information needed about applications, which compromises their applicability. In this paper, we propose a novel method for I/O management, IO-SETS. We present its potential through a scheduling heuristic called SET-10, which is simple and requires only minimal information. Our extensive experimental campaign shows the importance of IO-SETS and the robustness of SET-10 under various workloads. In particular in most of the simulated scenarios we improve the I/O slowdown over fairshare by 50%, which corresponds in our scenarios to a platform utilization gain of 2.5%. In the practical scenarios that we did, the utilization gain varies between 10 and 30%. We also provide insights on using our proposal in practice.

show abstract

“…We assume that the I/O nodes are fixed at one side of the mesh. These assumptions are motivated by our earlier measurements on an Intel TFLOPS [5,6] and by driving our simulations with a real workload trace taken from an Intel Paragon.…”

Section: Job Schedulingmentioning

confidence: 99%