John D. McCalpin
ABSTRACT
For the last decade, HPC systems have been dominated by clusters of two-socket commodity x86 servers, typically equipped with a non-commodity high-performance interconnect. Trends in lifecycle costs and prices, hardware technology, several measures of CPU and memory performance, and application performance characteristics are presented using several non-traditional perspectives. The evolution of the various "balances" of the systems over time is discussed --- both in the context of the interaction of application performance with the changing hardware, and in the context of the broader economic environment. Several serious obstacles to maintaining previous performance growth rates are identified and discussed, and it is argued that these are better viewed as architectural and market issues, rather than as fundamental technology issues. It is argued that overcoming these obstacles will require a fundamentally different approach to hardware architecture and programming languages, as well as to system configuration, deployment, and allocation strategies.
Recommendations
Intel Architecture and Technology for Future HPC System Building Blocks
ISPDC '15: Proceedings of the 2015 14th International Symposium on Parallel and Distributed ComputingIntel Corporation developed several new and enhanced technologies bolstering its leadership in high-performance computing.These new products include the future generation Intel Xeon Phi processor, code-named Knights Hill, and new high-speedinterconnect ...
Performance of the NVIDIA Jetson TK1 in HPC
CLUSTER '15: Proceedings of the 2015 IEEE International Conference on Cluster ComputingThe NVIDIA Jetson is demonstrated as a competitiveHPC platform. The Jetson has 192 Kepler CUDA cores that are"true" in that they share a processor: in the case of the Jetson, a32-bit ARM Cortex-A15 variant low power architecture. Ourwork explores the ...
Characterizing the performance benefit of hybrid memory system for HPC applications
Highlights- We perform memory-centric analysis of tested applications.
- We quantify the ...
AbstractHeterogenous memory systems that consist of multiple memory technologies are becoming common in high-performance computing environments. Modern processors and accelerators, such as the Intel Knights Landing (KNL) CPU and NVIDIA Volta ...
Comments