Acceleration-as-a-{\mu}Service: A Cloud-native Monte-Carlo Option Pricing Engine on CPUs, GPUs and Disaggregated FPGAs

Beat Weiss; Burkhard Ringlein; Christoph Hagleitner; Dionysios Diamantopoulos; Francois Abel; Mark Lantz; Mitra Purandare; Raphael Polig

arxiv: 2106.06293 · v1 · pith:KPNFQXZH · submitted 2021-06-11 · cs.NI · cs.AR· cs.DC

Acceleration-as-a-{μ}Service: A Cloud-native Monte-Carlo Option Pricing Engine on CPUs, GPUs and Disaggregated FPGAs

Dionysios Diamantopoulos , Raphael Polig , Burkhard Ringlein , Mitra Purandare , Beat Weiss , Christoph Hagleitner , Mark Lantz , Francois Abel This is my paper

Reviewed by Pithpith:KPNFQXZH open to challenge →

classification cs.NI cs.ARcs.DC

keywords cloudacceleratorsframeworkmicroservicesserviceworkloadacceleration-as-a-accelerator

0 comments

read the original abstract

The evolution of cloud applications into loosely-coupled microservices opens new opportunities for hardware accelerators to improve workload performance. Existing accelerator techniques for cloud sacrifice the consolidation benefits of microservices. This paper presents CloudiFi, a framework to deploy and compare accelerators as a cloud service. We evaluate our framework in the context of a financial workload and present early results indicating up to 485x gains in microservice response time.

This paper has not been read by Pith yet.

Acceleration-as-a-{μ}Service: A Cloud-native Monte-Carlo Option Pricing Engine on CPUs, GPUs and Disaggregated FPGAs

discussion (0)