A Simple QoS Packet Scheduler for Network Routers

1
A Simple QoS Packet Scheduler for Network Routers

• Su Wen
• Presenting the paper in SIGCOM 2001
• SRR An O(1) Time Complexity Packet Scheduler
  for Flows in Multi-Service Packet Networks

2
Quality of Service

• Current Internet Service
• Best effort
• Packet scheduling FIFO
• QoS
• Demanded by new network services
• multimedia, audio/video
• Meets requirement in bandwidth, delay, jitter

3
QoS Approaches

• Integrated Services
• Provide service guarantees to individual flows in
  the network
• GPS - Generalized Processor Sharing
• Weighted Fair Queuing
• Differentiated Services
• Provide some service guarantees to different
  classes of traffic in the network
• Example hierarchical link-sharing
• audio, video, telnet, ftp, mail

4
Bandwidth Sharing

• GPS - generalized processor sharing
• ideal model each flow gets its fair share of BW
• Si(t1, t2) / Sj(t1, t2) wi/wj
• ri/r wi/?jwj

flows

• wi - weight of flow i
• Si (t1, t2) - amount of traffic served for flow I
  during time t1 to t2
• ri - service rate of flow i
• r - service rate of the server

5
Bandwidth Sharing Implementations

• Weighted Fair Sharing
• An approximation/implementation of GPS
• Each flow has separate queues
• Two approaches
• time-stamp based e.g. virtual clock
• round robin

6
Implementation Characteristics

• Time Stamp based
• Fairly good fairness and delay bound
• Time complexity O(logN) - O(N)
• Round Robin based DRR, CORR
• Time complexity O(1)
• Short term unfairness
• bursty output

7
SRR - Smoothed Round Robin

• Good short term and long term fairness
• emulate GPS
• work conserving
• forwards packets as long as there are active
  flows
• O(1) time complexity
• one lookup to decide what to send
• Better scalability
• Simplicity

8
Data Structures

• Weight Matrix (WM)
• number of rows N - number of flows
• number of columns k log2wmax 1
• of binary digits needed to represent the max
  normalized weight
• WM size depends on of flows and BW granularity
• WSS - Weight Spread Sequence
• WSS of order k (Sk) is a sequence of integers
  from 1 to k
• recursively defined
• S1 1
• Sk Sk-1, k, Sk-1

9
Weight Matrix

• Four flow f1 - f4, each with different BW
  requirement
• r1 64kbps, r2 128kbps, r3 320kbps, r4
  192kbps
• w1 1 -gt 0 0 1
• w2 2 -gt 0 1 0
• w3 5 -gt 1 0 1
• w4 3 -gt 0 1 1
• col1 col2 col3
• row1 0 0 1
• row2 0 1 0
• row3 1 0 1
• row4 0 1 1

10
Weight Spread Sequence
Rule S1 1 Sk Sk-1, k, Sk-1

• To find WSS of order 3 - S3
• S2, 3, S2
• S1, 2, S1, 3, S1, 2, S1
• 1, 2, 1, 3, 1, 2, 1

N - number of flows wi - normalized weight of
flow i k - log2wmax 1
11
Scheduling Algorithm

• Generate a Service Queue
• for each value i in the WSS
• if jth row of column i of WM has value 1
• add flow j to the queue
• Serve the flow indicated in the service queue in
  a round robin fashion

f1 0 0 1 f2 0 1 0 f3 1 0 1 f4 0 1 1
S3
1, 2, 1, 3, 1, 2, 1
Flow Service Queue
f2, f4,
f3,
f2, f4,
f1, f3, f4,
f3
f3
f3,
12
Fairness

• Long Term
• Si(0, t) / Sj(0, t) wi/wj
• Si(0, t)/wi - Sj(0, t)/wj 0
• Short Term
• Si(0, t)/wi - Sj(0, t)/wj lt (k2)Lmax/2min(wi,wj)
  

Si(t1, t2) - amount of traffic served for flow i
during time t1 to t2 wi - normalized weight of
flow i k - log2wmax 1, the order of WSS
13
Delay

• Df - maximum delay for flow f
• Df lt 2Lmax/wf 2(N-1)Lmax/ ?i wi
• Inverse proportional to the weight of the flow
• Proportional to total number of flows
• Not strictly rate proportional, but better than
  DRR
• simulation result shows so

14
Scalability

• Space requirement
• N x k
• At 1kbps granularity, k16 can accommodate rates
  up to 64Mbps
• At 1Mbps granularity, k16 can accommodate rates
  up to 64Gbps
• Time complexity
• O(1) to choose packets for transmission
• O(k) to add or delete a flow

15
SRR Problems

• add_flow() and delete_flow() operations
• SRR scheduler is called during busy period
• packets are back-logged
• add_flow() is called when new flow arrives
• delete_flow() is called when queue of the flow is
  empty
• If a flow is not back logged, add_flow() and may
  be called for every packet
• quite expensive O(k)
• Solution
• delay the deletion of a flow
• not strictly O(1) any more

16
Conclusion

• SRR is simple
• even I can understand!
• Scalable
• time and space
• Good performance
• fairness
• delay

17
Questions?

• How did he come up with the algorithm?