Scalability! but at what cost?
2015
218 citations
Abstract
We offer a new metric for big data platforms, COST, or the Configuration that Outperforms a Single Thread. The COST of a given platform for a given problem is the hardware configuration required before the platform out-performs a competent single-threaded implementation. COST weighs a system’s scalability against the over-heads introduced by the system, and indicates the actual performance gains of the system, without rewarding sys-tems that bring substantial but parallelizable overheads. We survey measurements of data-parallel systems re-cently reported in SOSP and OSDI, and find that many systems have either a surprisingly large COST, often hundreds of cores, or simply underperform one thread for all of their reported configurations. 1
Keywords
ScalabilityParallelizable manifoldComputer scienceThread (computing)Distributed computingBig dataEmbedded systemParallel computingOperating system
Affiliated Institutions
Related Publications
TSAaaS: Time Series Analytics as a Service on IoT
In recent years, the evolving of IoT (Internet of Things) has resulted in the deployment of massive numbers of sensors in various fields, such as the Energy and Utility (E&U) in...
Efficient Architecture-Aware Acceleration of BWA-MEM for Multicore Systems
Innovations in Next-Generation Sequencing are enabling generation of DNA sequence data at ever faster rates and at very low cost. For example, the Illumina NovaSeq 6000 sequence...
Scalable molecular dynamics with NAMD
Abstract NAMD is a parallel molecular dynamics code designed for high‐performance simulation of large biomolecular systems. NAMD scales to hundreds of processors on high‐end par...
Publication Info
- Year
- 2015
- Type
- article
- Pages
- 14-14
- Citations
- 218
- Access
- Closed
External Links
Citation Metrics
218
OpenAlex
Cite This
Frank McSherry,
Michael Isard,
Derek G. Murray
(2015).
Scalability! but at what cost?.
, 14-14.