Introduction to Content-aware Switch

1
Introduction to Content-aware Switch

• Presented by
• Li Zhao

2
Content-aware Switch (CS)
www.yahoo.com
Internet
Image Server
APP. DATA
TCP
IP
Application Server
Switch
GET /cgi-bin/form HTTP/1.1 Host www.yahoo.com
HTML Server

• Front-end of a web cluster
• Route packets based on layer 5/7 (content)
  information

3
Why use CS

• Servers can be specialized for certain types of
  request
• Content segregation
• Exploit locality
• Affinity-based routing
• Increase the performance because of the improved
  hit rate
• Partial replication of server file set
• Partition the servers file set over different
  nodes

4
Content-aware Switch Architecture

• Two way architecture
• Server returns the
• response to the switch
• One way architecture
• Server returns the
• response to the client

server
switch
client
Valeria01
5
Layer 7 Two-way Architecture
6
Layer-7 Two-way Mechanisms

• TCP gateway
• An application level proxy running on the web
  switch mediates the communication between the
  client and the server
• TCP splicing
• reduce the overhead in TCP gateway. Packet
  forwarding occurs at network level between the
  network interface driver and the TCP/IP stack, is
  carried out directly by OS

user
kernel
user
kernel
7
TCP Splicing
client

server

content switch


SYN(CSEQ)
step1


step2

SYN(DSEQ)


ACK(CSEQ1)
DATA(CSEQ1)

step3

ACK(DSEQ1)
step7

DATA(SSEQ1)
DATA(DSEQ1)



ACK(CSEQlenR1)
ACK(CSEQLenR1)
step8


ACK(DSEQ
lenD1)

ACK(SSEQlenD1)

lenR size of http request.

.

lenD size of return document
8
TCP Splicing w/ Pre-forked Connections
switch
client
server
SYN(PSEQ)
step1
SYN(SSEQ)
step2
ACK(PSEQ1)
ACK(SSEQ1)



step3

SYN(CSEQ)


step4

SYN(DSEQ)

step5

ACK(CSEQ1)
DATA(CSEQ1)


step6
ACK(DSEQ1)
DATA(PSEQ1)


step7
ACK(SSEQ1)
DATA(SSEQ1)
DATA(DSEQ1)




step8
ACK(PSEQlenR1)
ACK(CSEQLenR1)





ACK(DSEQ
lenD1)

ACK(SSEQlenD1)

step9


lenR size of http request.


Ref Yang99
.

lenD size of return document
9
Pre-Allocate Server Scheme
Pre-allocated server
client


content switch

• Use a guess routing decision based on
  IP/Port/History
• Advantage
• Faster than TCP splicing.
• Reduce session processing overhead no need to
  convert server sequence

Ref Edward
10
Degenerated to TCP Splicing If Guess Wrong
Pre-allocated server
client


content switch


SYN(CSEQ)
SYN(CSEQ)
step1


SYN(SSEQ)
SYN(SSEQ)
step2




ACK(CSEQ1)

ACK(CSEQ1)
step3
DATA(CSEQ1)

FIN(CSEQ1)



ACK(SSEQ1)
Right server
step4

DATA(SSEQ1)

DATA(RSEQ1)
ACK(CSEQLenR1)
ACK(CSEQlenR1)

step5
lenD1)
ACK(DSEQ
ACK(SSEQlenD1)

Sequence conversion needed
11
Case Study

• Linux-based content aware switch Yang99
• IBM Layer 5 Pradhan00

12
Functional Overview of Content-aware Distributor
Ref Yang99
13
Results

• Overhead of the switch
• 89usec reduced ? pre-forked connections
• CS vs. Layer 4 switch
• Affinity-based routing vs. WRR
• Content-segregation vs. WRR
• CGI 27
• Static 36

14
IBM Switch Architecture

• Switch core
• Port controller
• Identify packets (layer 5) and send them to CPU
• Processing all other packets
• CPU PowerPC 603e
• Parse http request
• URL based routing

Ref Pradhan99
15
Flow Diagram on Layer 5 System

• Client ports vs. server ports
• Classifier Identify packets

16
Results

• CS vs. Layer 4 switch
• Entire set of files are replicated
• Some servers share files by NFS
• Partitioned file set

17
Layer-7 one-way architecture
18
Layer-7 one-way mechanisms

• TCP handoff
• The switch hands off the TCP connection endpoint
  to the server
• TCP connection hop
• Software-based proprietary solution
• encapsulating the IP packet in an RPX packet and
  sending it to the server.

19
TCP Handoff
client

content switch

server

SYN(CSEQ)
step1



step2
SYN(DSEQ)


ACK(CSEQ1)
DATA(CSEQ1)

step3
ACK(DSEQ1)
Migrate(Data, CSEQ, DSEQ)

step4




DATA(DSEQ1)
step5



ACK(CSEQlenR1)


step6

ACK(DSEQ
lenD1)

ACK(DSEQlenD1)

• Migrate the created TCP connection from the
  switch to the back-end sever
• Create a TCP connection at the back-end without
  going through the TCP three-way handshake
• Retrieve the state of an established connection
  and destroy the connection without going through
  the normal message handshake required to close a
  TCP connection
• Once the connection is handed off to the back-end
  server, the switch must forward packets from the
  client to the appropriate back-end server

Pai98
20
Case Study

• Scalable content-aware request distribution in
  cluster-based network servers Aron00

21
TCP Handoff
(1) a client connects to the front-end (2) the
dispatcher at the front-end accepts the
connection and hands it off to a back-end server
using the handoff protocol (3) the back-end takes
over the established connection received by the
handoff protocol (4) the server at the back-end
accepts the created connection (5) the server at
the back-end sends replies directly to the client
22
Scalability of a single Front-end
23
Scalable Cluster Design

• Switch
• Dispatcher component
• Implement the request distribution decide which
  server should handle request
• 0.8usec
• Distributor component
• Distribute the client requests to the server
  (handoff or splicing)
• 300usec for handoff, gt750usec for splicing

24
Cluster Operation

• (1) The layer 4 switch receives a SYN packet,
  choose the least loaded distributor
• (2) the distributor accepts the TCP connection
  and parses the client request
• (3) the distributor contacts the dispatcher for
  the assignment of the request to a server
• (4) the distributor hands off the connection
  using TCP handoff protocol to the server
• (5) the server takes over the connection using
  its handoff protocol
• (6) the server application at the server node
  accepts the connection
• (7) The server sends the response directly to the
  client
• (8) (not shown) the switch forward TCP
  acknowledgments to the corresponding server

25
Results

• The proposed cluster architecture scales far
  better than the one with a single front-end node.

26
Our Current Research on CS

• IXP 1200
• StrongARM _at_ 233MHz
• Microengine(6)
• IXP 2400
• Xscale _at_ 700MHz
• Microengines(8)

27
Our Design
28
Using TCP Splicing
29
Results
30
References

• Pradhan00 G.Apostolopoulos, et. al, Design,
  Implementation and Performance of a Content-Based
  Switch, proceedings of IEEE INFOCOM-2000
• Pai98 V.S. Pai, et. al, Locality-Aware Request
  Distribution in Cluster-based Network Servers. In
  Proceedings of the 8th Conference on
  Architectural Support for Programming Languages
  and Operating Systems, San Jose, CA, Oct.1998
• Aron00 Mohit Aron et. al, Scalable
  Content-aware Request Distribution in
  Cluster-based Network Servers, Proc. of the 2000
  Annual Usenix Technical Conference, June 2000
• Edward C. Edward Chow Chow, Introduction to
  content switch
• Valeria01 Valeria Cardellini, et. al, The state
  of the Art in Locally Distributed Web-server
  Systems, IBM research report
• Yang99 Chu-Sing Yang, et. Al, Efficient support
  for content-based rouging in web server clusters,
  Proc. Of USITS 99