Workloads for Programmable Network Interfaces - - PowerPoint PPT Presentation

workloads for programmable network interfaces
SMART_READER_LITE
LIVE PREVIEW

Workloads for Programmable Network Interfaces - - PowerPoint PPT Presentation

Workloads for Programmable Network Interfaces


slide-1
SLIDE 1

Workloads for Programmable Network Interfaces

slide-2
SLIDE 2

Motivation

slide-3
SLIDE 3

The old …

I/O Bus PCI RAM RAM CNTRL NET CPU

network

slide-4
SLIDE 4

… and the current

Myrinet 33Mhz Lanai IOP Intel PRO/100 Smart Adapter 66Mhz i960 IOP Alteon ACEnic Gigabit Ethernet 2 100Mhz MIPS IOPs

I/O Bus PCI RAM RAM IOP NET

~1Gbps 128KB- 16MB

CPU

slide-5
SLIDE 5

Workloads

slide-6
SLIDE 6

Workloads

Applications Description Packet Classification/Filtering Claim/forward/drop decisions, statistics gathering, and firewalling. IP Packet Forwarding Forward IP packets based on routing information. Network Address Translation Translate between globally routable and private IP packets. Useful for IP masquerading, virtual web server, etc. TCP connection management Traffic shaping within the network to reduce congestion. TCP/IP Offload TCP/IP processing from Internet/Web servers. Web Switching Web load balancing and proxy cache monitoring. Virtual Private Network (VPN) IP Security (IPSec) Encryption (DES) and Authentication (MD5) Data Transcoding Converting a multimedia data stream from one format to another within the network. Duplicate Data Suppression Reduce superfluous duplicate data transmission over high cost links.

slide-7
SLIDE 7

Benchmark Characteristics

Application Insts Executed per M essage Loads/Stores (% ) Ctrl Flow (% ) O ther (% ) ip4lookup 120 18.6 12.1 69.3 M D5 23K 10.3 0.6 89.1

slide-8
SLIDE 8

Experimental Architectures

slide-9
SLIDE 9

Superscalar Performance

ip4lookup

0.0E+00 5.0E+06 1.0E+07 1.5E+07 2.0E+07 2.5E+07 1 2 3 4 5 6 7 8

# of functional units message per second

1T@500MHz 1T@400MHz 1T@300MHz 1T@200MHz 1T@100MHz

1Gbps 10Gbps

x

MD5

0.0E+00 1.0E+04 2.0E+04 3.0E+04 4.0E+04 5.0E+04 6.0E+04 7.0E+04 8.0E+04 9.0E+04 1 2 3 4 5 6 7 8

# of functional units message per second 1Gbps 100Mbps

slide-10
SLIDE 10

FGMT Performance

MD5

0.0E+00 1.0E+04 2.0E+04 3.0E+04 4.0E+04 5.0E+04 6.0E+04 7.0E+04 8.0E+04 9.0E+04 1 2 3 4 5 6 7 8

# of FGMT thread contexts

messages per second

100Mbps 1Gbps

ip4lookup

0.0E+00 5.0E+06 1.0E+07 1.5E+07 2.0E+07 2.5E+07 1 2 3 4 5 6 7 8

# of FGMT thread contexts messages per second

FGMT@500mhz FGMT@400mhz FGMT@300mhz FGMT@200mhz FGMT@100mhz

10Gbps 1Gbps

slide-11
SLIDE 11

SMT Performance

  • ip4lookup

0.0E+00 5.0E+06 1.0E+07 1.5E+07 2.0E+07 2.5E+07 3.0E+07 3.5E+07 4.0E+07 4.5E+07 5.0E+07 1 2 3 4 5 6 7 8

# of SMT thread contexts message per second

SMT@500mhz SMT@400mhz SMT@300mhz SMT@200mhz SMT@100mhz

1Gbps 10Gbps

MD5

0.0E+00 2.0E+04 4.0E+04 6.0E+04 8.0E+04 1.0E+05 1.2E+05 1.4E+05 1.6E+05 1.8E+05 1 2 3 4 5 6 7 8

# of SMT thread contexts messages per second

1Gbps 100Mbps

slide-12
SLIDE 12

Conclusions

slide-13
SLIDE 13

Future Work