Network vs. Server Issues in endtoend performance

1
Network vs. Server Issues in end-to-end
performance

• Christian Huitema
• huitema_at_microsoft.com
• June 18, 2000

2
The Internet grows, and grows
Number of hosts (millions) in the Internet
(Netsizer by Telcordia)
3
Network versus servers who is trailing whom?

• Network vs. server performance, today
• State of the network,
• The special case of the DNS,
• The 10 Gigabit challenge,
• Going end-to-end?

4
Network vs. server performance, today
5
Measuring the components of the web delay
get address
DNS
connect
delay
accept
HTTP GET
prepare
server
receive
transmit
network
close
Get address only once (cache).
Connect only once (HTTP 1.1), Pipeline.
6
Web performance estimation, 1998-2000

• For a given client/server pair
• Measure the address resolution delay, connection
  delay, delay to first packet, delay to
  transmission of page
  
• Repeat every day
• For a fixed set of 100 large servers,
• For a set of 100 random servers, provided from
  randomized requests to Altavista, Lycos, Google,
  etc.

7
Observed bandwidth seems to grow over time
8
The servers seem to contribute 40 of the delay
9
Feedback Loop Why Capacity Doubles Every 8
Months
More contents, more servers, faster feeds
Providers increase the network capacity
More users, more demand, faster modems
10
Network vs. server what is the state of the
network?
11
First network characteristicPacket loss rates
12
Second characteristicRound trip time
13
Taking a closer look at the round trip time
distribution
14
Likely evolution of the network

• Accommodate growing demand
• Cable, DSL, other broadband
• Mobile phones, appliances
• Likely evolution
• Average bandwidth grows - 100 kbps
• Median delays - distance/speed of light
• Delay deviation diminishes, as predicted by Paul
  Kennys work.

15
Network vs. serverThe special case of the DNS
16
DNS an amplifier of the Internet Performance
Root, .Com
Local Server
Example.Com
Many network traversals, many servers.
17
Resolution Delay,Random DNS Names
18
DNS delays dont seem to improve at Internet
Speed
Fraction of DNS resolution delays over 2 seconds
19
A clear case of server vs. network

• A.ROOT-SERVERS.NET loss 33.9
• B.ROOT-SERVERS.NET loss 3.4
• C.ROOT-SERVERS.NET loss 21.1
• D.ROOT-SERVERS.NET loss 2.6
• E.ROOT-SERVERS.NET loss 12.3
• F.GTLD-SERVERS.NET loss 4.0
• F.ROOT-SERVERS.NET loss 12.3
• G.ROOT-SERVERS.NET loss 62.1
• H.ROOT-SERVERS.NET loss 20.9
• I.ROOT-SERVERS.NET loss 7.1
• J.GTLD-SERVERS.NET loss 8.3
• K.GTLD-SERVERS.NET loss 2.2

Limits of these data One single test, One sing
le day!
20
How will the DNS situation evolve?

• No clear virtuous cycle
• Registration of new names is a for profit venture
  (NSI, and wannabes)
  
• Maintenance of root is a non profit service with
  political constraints (ICANN)
  
• Solutions from servers, software
• Less reliance on root?
• More reliance on caches (load balancing?)
• Less reliance on the DNS!

21
Network vs. serverThe 10 Gigabit challenge
22
Servers are getting better but is this enough?

• Internet2 Land Speed Record award
• "We hope this competition gets people thinking
  about enabling really revolutionary Internet
  applications," - Jim Gray
  
• "To realize Internet's full potential, end-to-end
  network performance needs to take a huge leap
  forward," - Gordon Bell
  
• The records
• 751.362 Mbps over 5,626 km (single TCP)
• 957.369 Mbps over 5,626 km (2 TCP)

23
The 10 Gigabit challenge, or the limits of
Moores law
24
Meeting the challengepart 1, ever better servers

• Get stronger
• Faster CPU, better memory architecture
• Faster I/O
• Get smarter
• HTTP 1.1, TCP, IPv6
• Better protocol design(s)
• Better software (doing that)
• Get help
• Offload packet processing (done that)
• Offload encryption (done for IPSEC)
• Offload key management (uh?)

25
Meeting the challengepart 2, larger clusters

• Servers x2 every 18 month.
• Can we manage a cluster of 100,000 servers?
• Load balancing, Software upgrade, Failover
• Can we develop distributed applications?

26
Meeting the challengepart 3, distribution
networks

• Push the content near the consumers
• Is this needed in a 10 Gigabit world?
• Does it work for more than GIF files?

27
Evolution of server, server farms and distribution

• Handling variability
• Heavy tail distribution of demand means flash
  crowds, etc.
  
• Could incite to a market of servers, or to
  adaptive servers.
  
• Pushing up hill
• Need a serious business case for investment in
  new servers
  
• Constantly testing the limits of the technology

28
Network vs. serverGoing end-to-end?
29
End to end distributionNAPSTER, Gnutella
Rendezvous, directory
30
End to end communication real time audio, video,
games
SIP proxy
SIP proxy
31
Meeting the challengedistribute the load!

• The return of the end-to-end argument
• Each new client brings its own resource
• The only known way of scaling
• Only place in the servers the functions that
  cannot possibly placed in the client!

32
Network vs. serverThank You!