From WikiChip
Difference between revisions of "nec/microarchitectures/sx-aurora"
< nec

(Die)
(Memory Hierarchy)
Line 38: Line 38:
  
 
== Memory Hierarchy ==
 
== Memory Hierarchy ==
{{empty section}}
+
* Vector core
 +
** SPU
 +
*** L1I Cache:
 +
**** 32 KiB
 +
*** L1D Cache:
 +
**** 32 KiB
 +
*** L2 Cache:
 +
**** 256 KiB
 +
** VPU
 +
*** 120 KiB [[load buffer]]
 +
*** 64 KiB [[store buffer]]
 +
* L3 Cache/LLC:
 +
** 16 MiB
 +
*** 8 x 2 MiB
 +
*** [[write-back]]
 +
*** inclusive of L1 & L2
 +
*** 128 banks
 +
*** 3 TiB/s bandwidth
  
 
== Overview ==
 
== Overview ==

Revision as of 00:26, 24 November 2018

Edit Values
SX-Aurora µarch
General Info
Arch TypeVPU
DesignerNEC
ManufacturerTSMC
Introduction2018
Core Configs8
Pipeline
TypeSuperscalar, Pipelined
OoOEYes
SpeculativeYes
Reg RenamingYes
Stages8
Decode4-way
Cache
L1I Cache32 KiB/core
L1D Cache32 KiB/core
L2 Cache256 KiB/core
L3 Cache16 MiB/chip
Succession

SX-Aurora is NEC's successor to the SX-ACE, a 16 nm microarchitecture for vector processors first introduced in 2018.

History

New text document.svg This section is empty; you can help add the missing info by editing this page.

Architecture

Key changes from SX-ACE

New text document.svg This section is empty; you can help add the missing info by editing this page.

Block Diagram

New text document.svg This section is empty; you can help add the missing info by editing this page.

Memory Hierarchy

  • Vector core
  • L3 Cache/LLC:
    • 16 MiB
      • 8 x 2 MiB
      • write-back
      • inclusive of L1 & L2
      • 128 banks
      • 3 TiB/s bandwidth

Overview

Overview of the SX-Aurora

The SX-Aurora is NEC's successor to the SX-ACE, a vector processor designed for high-performance scientific/research applications and supercomputers. The SX-Aurora deviates from all prior chips in the kind of markets it's designed to address. Therefore, NEC made slightly different design choice compared to prior generations of vector processors. In an attempt to broaden their market, NEC extended beyond supercomputers to the conventional server and workstation market. This is done through the use of PCIe-based accelerator cards.

Moving to an accelerator card is not without its challenges. To keep the high memory bandwidth, and thus high bytes per FLOP, while moving a smaller form factor, it was necessary to drop the large amount of DDR memory channels. Instead, NEC opted to utilizing high-bandwidth memory on-chip instead. The card itself is designed communicate with other cards on the system in order to scale up from just a single card for workstation use to a supercomputer with 64 cards per rack.

The chip itself consists of eight very big cores along with 16 MiB of last level cache on a 2-dimensional mesh. Attached to the LLC are the two memory controllers which interface with the six high-bandwidth memory sitting on an interposer. Fabricated on TSMC's 16 nm process, the SX-Aurora operates at up to 1.6 GHz delivering up to 307.2 gigaFLOPS (double-precision) per core for a total of up to 2.45 teraFLOPS.

Vector engine (VE) card

New text document.svg This section is empty; you can help add the missing info by editing this page.

Die

  • 16 nm process
  • 4,800,000,000 transistors
  • 14.96 mm x 33.00 mm
    • 493.68 mm² die size

Bibliography

  • Template:hcbib
  • Supercomputing 2018, NEC Aurora Forum
  • Some information was obtained directly from NEC