Slide 1
Slide 1 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Faster webapp
Grégoire Charvet
[email protected]
April 30, 2014
https://speakerdeck.com/geekingfrog/faster-webapp-understand-the-network
Grégoire [email protected] Understand the network April 30, 2014 1 / 65
Slide 2
Slide 2 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Overview
1 Networking 101
UDP
TCP
TLS
2 HTTP
3 Final thoughts
Grégoire [email protected] Understand the network April 30, 2014 2 / 65
Slide 3
Slide 3 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Disclaimer
Go read this book −→
Everything and much more is
in here.
Grégoire [email protected] Understand the network April 30, 2014 3 / 65
Slide 4
Slide 4 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Intro
Grégoire [email protected] Understand the network April 30, 2014 4 / 65
Slide 5
Slide 5 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Speed is a feature
Faster sites lead to:
better user retention
better user engagement
higher conversion
Grégoire [email protected] Understand the network April 30, 2014 5 / 65
Slide 6
Slide 6 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Two critical components
Figure: Speed and Latency
Grégoire [email protected] Understand the network April 30, 2014 6 / 65
Slide 7
Slide 7 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Route Distance Time, light
in a vac-
uum
Time, light
in a fiber
Taipei to Tokyo 2115 km 7 ms 10 ms
Taipei to Paris 9845 km 32 ms 49 ms
Taipei to SF 10346 km 34 ms 51 ms
Equatorial cir-
cumference
40 075 km 133 ms 200 ms
Table: Some latencies
Rule of thumb: speed of light in fiber: 2.108m/s
Grégoire [email protected] Understand the network April 30, 2014 7 / 65
Slide 8
Slide 8 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Measure latency
traceroute (linux)
tracert (windows)
$> traceroute google.com
traceroute to google.com (74.125.224.102), 64 hops max, 52 byte packets
1 10.1.10.1 (10.1.10.1) 7.120 ms 8.925 ms 1.199 ms
2 96.157.100.1 (96.157.100.1) 20.894 ms 32.138 ms 28.928 ms
3 x.santaclara.xxxx.com (68.85.191.29) 9.953 ms 11.359 ms 9.686 ms
4 x.oakland.xxx.com (68.86.143.98) 24.013 ms 21.423 ms 19.594 ms
5 68.86.91.205 (68.86.91.205) 16.578 ms 71.938 ms 36.496 ms
6 x.sanjose.ca.xxx.com (68.86.85.78) 17.135 ms 17.978 ms 22.870 ms
7 x.529bryant.xxx.com (68.86.87.142) 25.568 ms 22.865 ms 23.392 ms
8 66.208.228.226 (66.208.228.226) 40.582 ms 16.058 ms 15.629 ms
9 72.14.232.136 (72.14.232.136) 20.149 ms 20.210 ms 18.020 ms
10 64.233.174.109 (64.233.174.109) 63.946 ms 18.995 ms 18.150 ms
11 x.1e100.net (74.125.224.102) 18.467 ms 17.839 ms 17.958 ms
Grégoire [email protected] Understand the network April 30, 2014 8 / 65
Slide 9
Slide 9 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Two barriers
300ms: ”There is some lag”
1s: feels slugish ← upper limit to aim for.
Grégoire [email protected] Understand the network April 30, 2014 9 / 65
Slide 10
Slide 10 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Latency matter more than bandwidth
Figure: Bandwidth vs latency
Grégoire [email protected] Understand the network April 30, 2014 10 / 65
Slide 11
Slide 11 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
UDP
User datagram protocol
Grégoire [email protected] Understand the network April 30, 2014 11 / 65
Slide 12
Slide 12 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Simplest protocol - UDP
Super simple
Grégoire [email protected] Understand the network April 30, 2014 12 / 65
Slide 13
Slide 13 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
A few shortcomings
No state
No acknowledgement
No congestion control
No message splitting
No order delivery garantee
Grégoire [email protected] Understand the network April 30, 2014 13 / 65
Slide 14
Slide 14 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
TCP
Grégoire [email protected] Understand the network April 30, 2014 14 / 65
Slide 15
Slide 15 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
TCP (transfer control protocol)
Address the issues of UDP
Building block of the web
Grégoire [email protected] Understand the network April 30, 2014 15 / 65
Slide 16
Slide 16 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Three-Way handshake
Figure: Three-Way handshake
Grégoire [email protected] Understand the network April 30, 2014 16 / 65
Slide 17
Slide 17 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Flow control
receive window (rwnd) to control the size of buffers
Slow start & Congestion avoidance
Grégoire [email protected] Understand the network April 30, 2014 17 / 65
Slide 18
Slide 18 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Receive window
Figure: Receive window advertisement
Grégoire [email protected] Understand the network April 30, 2014 18 / 65
Slide 19
Slide 19 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Slow start
Avoid congestion collapse by starting sending few packets
Measure available bandwidth by sending packet
Dropped packet (missed ACK) trigger adjustements of cwnd
Grégoire [email protected] Understand the network April 30, 2014 19 / 65
Slide 20
Slide 20 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Figure: Packet dropped
Grégoire [email protected] Understand the network April 30, 2014 20 / 65
Slide 21
Slide 21 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Slow start and congestion avoidance
Automatic control of the congestion window
Grégoire [email protected] Understand the network April 30, 2014 21 / 65
Slide 22
Slide 22 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Time required to reach maximum throughput
Time to reach a cwnd of size N;
Time = RTT ×
⌈
log
2
(
N
init cwnd
)⌉
Grégoire [email protected] Understand the network April 30, 2014 22 / 65
Slide 23
Slide 23 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Example
Client and server rwnd: 65535 bytes (64K)
Initial congestion window: 4 segments (RFC 2581)
Roundtrip time: 56 ms (London to New York)
Size of one segment: 1460 bytes (MTU - ipv4 overhead)
Number of segments required to get to 64K:
65 535 bytes
1460 bytes
≈ 45 segments
Time for cwnd to reach 45 segments:
56 ms ×
⌈
log
2
(
45
4
)⌉
= 224 ms
Grégoire [email protected] Understand the network April 30, 2014 23 / 65
Slide 24
Slide 24 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Example cont’d
Figure: Slow start and congestion window growth
Grégoire [email protected] Understand the network April 30, 2014 24 / 65
Slide 25
Slide 25 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Implications
FOUR round trip !
Bandwidth is irrelevant here
Big problem for most HTTP connection (bursty traffic)
Grégoire [email protected] Understand the network April 30, 2014 25 / 65
Slide 26
Slide 26 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Fetching a file over a new TCP connection
Fetch a file of 40
KB from New-York
to London.
Grégoire [email protected] Understand the network April 30, 2014 26 / 65
Slide 27
Slide 27 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Reuse of tcp connection
Getting the same file over the same connection.
Figure: Same file over the same connection
Grégoire [email protected] Understand the network April 30, 2014 27 / 65
Slide 28
Slide 28 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Summary of TCP
Three-Way handshake → full RTT of latency
Slow start applied for every new connections
Flow and congestion control regulate throughput
Congestion window size control throughput
Grégoire [email protected] Understand the network April 30, 2014 28 / 65
Slide 29
Slide 29 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
TCP optimizations
TCP fastopen: send data with the first syn packet (linux kernel
≥ 3.7)
Window scaling: allow greater receive window
disable slow start restart
$> sysctl net.ipv4.tcp_slow_start_after_idle
$> sysctl -w net.ipv4.tcp_slow_start_after_idle=0
Increase initial congestion window size to 10 (kernel ≥ 2.6.39)
Reuse tcp connection
Put server close to clients
Compress transferred data
Grégoire [email protected] Understand the network April 30, 2014 29 / 65
Slide 30
Slide 30 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
TLS
Grégoire [email protected] Understand the network April 30, 2014 30 / 65
Slide 31
Slide 31 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
TLS or SSL ?
Original name: SSL, developed at Netscape.
Renamed to TLS when standardized by the IETF.
Grégoire [email protected] Understand the network April 30, 2014 31 / 65
Slide 32
Slide 32 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Encryption, Authentication, Integrity
TLS provides:
Encryption
Authentication
Integrity
As a side effect: protects whatever application is running on top of it
from misconfigured middle agent (proxies).
Grégoire [email protected] Understand the network April 30, 2014 32 / 65
Slide 33
Slide 33 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Computational cost
“SSL/TLS is not computationally expensive anymore.”
— Adam Langley (Google) - 2010
We have found that modern software-based TLS implementations
running on commodity CPUs are fast enough to handle heavy
HTTPS traffic load without needing to resort to dedicated
cryptographic hardware. We serve all of our HTTPS traffic using
software running on commodity hardware.
— Doug Beaver (Facebook)
Grégoire [email protected] Understand the network April 30, 2014 33 / 65
Slide 34
Slide 34 text
TLS handshake
Figure: TLS handshake
. . . .... .... .... . . . . .
Slide 35
Slide 35 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Optimizing TLS
TLS session resumption
session ticket
TLS false start
Grégoire [email protected] Understand the network April 30, 2014 35 / 65
Slide 36
Slide 36 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Session resumption
Re-use an already negotiated key.
Avoid a full round-trip.
A must have in most circumstances.
Problem on the servers: must maintain a cache for each session
ID.
Even more tricky when it comes to multiple servers sharing the
same cache.
Grégoire [email protected] Understand the network April 30, 2014 36 / 65
Slide 37
Slide 37 text
Session resumption cont’d
Figure: Abbreviated handshake
. . . .... .... .... . . . . .
Slide 38
Slide 38 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Session ticket
Encrypt all session’s data and store it on the client.
Only the server knows the key.
No server side caching.
But servers must be initialized with the same secret key (multi
server behind load balancer concerns). Also need structure to
deploy new keys
Grégoire [email protected] Understand the network April 30, 2014 38 / 65
Slide 39
Slide 39 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
False start
What about new visitors ?
Client can send data as soon at it knows the session key.
Grégoire [email protected] Understand the network April 30, 2014 39 / 65
Slide 40
Slide 40 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
False start
Figure: TLS false start
Grégoire [email protected] Understand the network April 30, 2014 40 / 65
Slide 41
Slide 41 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Compatibility of false start
There are some issue (refer to the book for details).
Grégoire [email protected] Understand the network April 30, 2014 41 / 65
Slide 42
Slide 42 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
TLS record
Figure: TLS record
Record can be split in multiple TCP packets.
Record must be complete to decrypt the data.
Grégoire [email protected] Understand the network April 30, 2014 42 / 65
Slide 43
Slide 43 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Optimizing TLS record size
IP overhead: 20 bytes (v4) or 40 bytes (v6)
TCP overhead: 20 bytes
TCP options overhead: 40 bytes
usual MTU: 1500 bytes.
Initial size, when to upgrade to what size ?
Grégoire [email protected] Understand the network April 30, 2014 43 / 65
Slide 44
Slide 44 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Google’s implementation
As early 2014: TLS record set to fit into one TCP packet for the first
MB, then upgraded to 16KB.
Reset to initial value after ≈ 1s of inactivity.
Grégoire [email protected] Understand the network April 30, 2014 44 / 65
Slide 45
Slide 45 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
TLS compression
Don’t do it.
CRIME attack
will compress everything (including image and other already
compressed stuff)
Grégoire [email protected] Understand the network April 30, 2014 45 / 65
Slide 46
Slide 46 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
HTTP
Grégoire [email protected] Understand the network April 30, 2014 46 / 65
Slide 47
Slide 47 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Rapid history
HTTP 0.9
Limited to html and ascii.
Connection closed after response.
HTTP 1.0
Introduced status code, request and response headers.
Response is not limited to html.
Connection is closed after response.
HTTP 1.1
Connection re-use by default
Complex, tons of options (content-encoding, caching directive,
language negotiation and so on...)
HTTP 2.0 (spdy)
Work in progress. SPDY is a working implementation.
Grégoire [email protected] Understand the network April 30, 2014 47 / 65
Slide 48
Slide 48 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Web application requests
Figure: One http request
Grégoire [email protected] Understand the network April 30, 2014 48 / 65
Slide 49
Slide 49 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Browser optimizations
Modern browsers do a lot of work under the hood to provide better
experience.
DNS pre-resolve and TCP pre-connect.
Speculative optimization.
Resources pre-fetching and prioritization.
Grégoire [email protected] Understand the network April 30, 2014 49 / 65
Slide 50
Slide 50 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Helping the browsers
1
2
3
4
1: pre-resolve hostname.
2: pre-fetch critical resource used later in the page.
3: pre-fetch resource for future navigation.
4: pre-render page for future navigation.
Grégoire [email protected] Understand the network April 30, 2014 50 / 65
Slide 51
Slide 51 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Browser support
Browser dns-prefetch subresource prefetch prerender
Firefox 3.5+ n/a 3.5+ n/a
Chrome 1.0+ 1.0+ 1.0+ 13+
Safari 5.01+ n/a n/a n/a
IE 9+ (prefetch) n/a 10+ 11+
Table: Speculative optimization support
Grégoire [email protected] Understand the network April 30, 2014 51 / 65
Slide 52
Slide 52 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
HTTP optimizations
HTTP 0.9 or 1.0 → upgrade to 1.1
Reduce DNS lookup
Fewer HTTP requests
Use a CDN
Use HTTP caching mechanisms
Gzip assets
Avoid HTTP redirect (even worse if it redirect to a different
host)
Grégoire [email protected] Understand the network April 30, 2014 52 / 65
Slide 53
Slide 53 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Figure: Benefit of connection keep-alive
Grégoire [email protected] Understand the network April 30, 2014 53 / 65
Slide 54
Slide 54 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
HTTP pipelining
Grégoire [email protected] Understand the network April 30, 2014 54 / 65
Slide 55
Slide 55 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Pipelining concerns
Clients and server must support it.
Application must handle aborted connections.
Must protect from broken intermediaries (use HTTPS)
Head of line blocking
The head of line blocking problem actually makes pipelining difficult
to use in practice. Browsers open 6 TCP connections to a server and
process requests in parallel.
Grégoire [email protected] Understand the network April 30, 2014 55 / 65
Slide 56
Slide 56 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Multiple TCP connections
In practice, browsers open 6 TCP connections per hosts.
Why 6? Experimentation.
No head of line blocking
More bytes can be sent within the first cwnd
But
More complexity to manage the sockets collections
More memory and cpu
Limited application parallelism
Resource exhaustion
Grégoire [email protected] Understand the network April 30, 2014 56 / 65
Slide 57
Slide 57 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Figure: Resource exhaustion
Grégoire [email protected] Understand the network April 30, 2014 57 / 65
Slide 58
Slide 58 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Domain sharding
Put resources on different domain (pic1.example.com,
pic2.example.com)
Cost: new dns request, new tcp connections (slow start)
Measure it (consider sharding when you have more than 10
resources download simultanously)
Grégoire [email protected] Understand the network April 30, 2014 58 / 65
Slide 59
Slide 59 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
HTTP optimizations, application level
Concatenation & spritting (less requests)
resource inlining (+33% size for images)
Grégoire [email protected] Understand the network April 30, 2014 59 / 65
Slide 60
Slide 60 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Final thoughts
Grégoire [email protected] Understand the network April 30, 2014 60 / 65
Slide 61
Slide 61 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
Optimization checklist
Always do:
Reduce dns lookups.
Reuse tcp connections.
Minimize number of HTTP request.
Use a cdn.
Eliminate unecessary resources and transferred bytes.
Compress data during transfer.
Cache resources on the client.
Parallelize resources fetching.
Grégoire [email protected] Understand the network April 30, 2014 61 / 65
Slide 62
Slide 62 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
HTTP 1.1 optimizations
HTTP pipelining (if you control both client and server this is
viable).
Domain sharding (warnings apply).
Bundle resources (less requests).
Inline small resources.
Grégoire [email protected] Understand the network April 30, 2014 62 / 65
Slide 63
Slide 63 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
What’s coming next ?
HTTP2 and SPDY
WebRTC
Grégoire [email protected] Understand the network April 30, 2014 63 / 65
Slide 64
Slide 64 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
And remember: measure then tweak
Grégoire [email protected] Understand the network April 30, 2014 64 / 65
Slide 65
Slide 65 text
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
The end
Grégoire [email protected] Understand the network April 30, 2014 65 / 65