We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.AR

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Hardware Architecture

Title: The nanoPU: Redesigning the CPU-Network Interface to Minimize RPC Tail Latency

Abstract: The nanoPU is a new networking-optimized CPU designed to minimize tail latency for RPCs. By bypassing the cache and memory hierarchy, the nanoPU directly places arriving messages into the CPU register file. The wire-to-wire latency through the application is just 65ns, about 13x faster than the current state-of-the-art. The nanoPU moves key functions from software to hardware: reliable network transport, congestion control, core selection, and thread scheduling. It also supports a unique feature to bound the tail latency experienced by high-priority applications. Our prototype nanoPU is based on a modified RISC-V CPU; we evaluate its performance using cycle-accurate simulations of 324 cores on AWS FPGAs, including real applications (MICA and chain replication).
Comments: 10 pages
Subjects: Hardware Architecture (cs.AR); Networking and Internet Architecture (cs.NI)
ACM classes: C.1.1; C.2.1
Cite as: arXiv:2010.12114 [cs.AR]
  (or arXiv:2010.12114v1 [cs.AR] for this version)

Submission history

From: Stephen Ibanez [view email]
[v1] Fri, 23 Oct 2020 00:42:40 GMT (1454kb,D)

Link back to: arXiv, form interface, contact.