NetBench: A Benchmarking Suite for Network Processors

1
NetBench A Benchmarking Suite for Network
Processors

• Wendong Hu
• Gokhan Memik
• Department of Electrical Engineering
• University of California, Los Angeles

2
Overview

• What is inside a router?
• A buyer wants the best
• Designer would like to build the best
• But what is the best?
• Benchmarks
• SPEC, TPC, SPLASH,

3
Outline

• Introduction
• Network Processor Survey
• Programs
• Results
• Conclusion

4
Introduction

• What is a Network Processor?
• Programmable ICs based on system-on-a-chip
  technology that perform communications-specific
  functions more efficiently than general-purpose
  processors

NP
5
Introduction (cont.)

• Why Network Processor(NP)
• 1 Gbps is limit for off-the shelf processors
• Emerging technologies and applications
• What did we do
• Survey NP
• Decide on a set of applications
• Who and why are people going to use NPs
• Gather statistics and compare

6
Companies

• Acorn Networks (genFlow) Alchemy (Au1000)
  AMCC/MMC (nP) Bay Micro BOPS (Manta)
  Broadcom/SiByte (SB-1 core SB-1250) Chameleon
  (CS2112) Chrysalis-ITS Cisco (Toaster 2)
  Cognigine Conexant/Maker (CX27510, MXT4400)
  empowerTel (MxP) Extreme Packet Devices EZChip
  Fast-Chip Hi/fn Hyperchip IBM (PowerNP) IDT
  Improv (Jazz) Infineon Intel/Level One
  (IXP1200) Internet Machines Lexra (NetVortex IP
  core) Lucent/Agere (Payload Plus) MIPS
  Motorola/C-Port (C-5) Philips PixelFusion
  Pluris PMC-Sierra/QED Redback Networks/Siara
  Silicon Access Networks (iFlow) Solidum
  Stargate StarGen TranSwitch T.sqware Vertex
  Networks (XpressFlow) Vitesse/Sitera (Prism
  IQ2000) Vitesse/XaQti (Active-Flow Processor)
  XStream ZettaCom (ZEN).

7
Example Design
8
Example Design
9
Applications

• Web applications
• Load-balancing, URL-based switching, web caching
• Security
• firewall, VPN, access control, encryption
• Routing and Switching
• VoIP gateways, IP routing, Smart routing
  mechanisms (DRR?)
• QoS issues
• MPLS, QoS scheduling and/or classification

10
Programs in NetBench

• Programs from 3 different levels
• Micro-codes
• Low-level tasks (CRC, memcopy, )
• IP level programs
• Routing and related applications
• High level programs
• Encryption-decryption, load balancing, Level 7
  switching

11
Complete list

• High-level
• UBS (URL-based switching)
• RTS (Real-Teal Streaming)
• TCS4 (load balancing)
• DF (Diffie-Hellman)
• MD5 (Message Digest)
• Micro-level
• CRC-32
• Table-lookup

12
Complete list (cont.)

• IP level
• MCAST (multicasting)
• DRR (Deficit Round Robin)
• CEDF (Coordinated earliest deadline first)
• NAT (Network address translation)
• IPv6TB (IP version 6 tunneling)
• IPCHAINS (firewall)
• IPATM (IP over ATM support)
• Route (IPv4 routing)

13
Methodology

• Implementations
• Use the available software
• Not too optimized for a specific architecture
• Some applications from FreeBSD
• SimpleScalar Simulator
• Most popular simulator in Architecture community
  (cycle-based)
• Extensive information about program execution
• Intel IXP1200 Simulator
• Micro-code assembler and debugger
• Accurate, detailed cycle-based simulator

14
Input Trace

• From Columbia University

15
Applications in SimpleScalar

• MD5
• DRR
• NAT
• Route
• Table Lookup
• CRC_32

16
Execution Times
17
Cache behaviour

• Miss rates are small compared to SPEC

18
Parallelism

• IPC Instruction executed per cycle

19
Routing Table Size
20
Intel IXP1200
21
Conclusion and Future Work

• Need innovative architectures for fast link and
  emerging applications
• Network processors deliver the necessary
  flexibility and processing power
• Need a framework to compare the performances gt
  NetBench
• Some applications are not as tough as predicted
• Continue porting applications and running
  simulations