1. Packet Classification Using Binary Content Addressable Memory
2019/11/2
Packet Classification Using Binary Content Addressable Memory
Presenter: Wei-Li,Wang
IEEE/ACM TRANSACTIONS ON NETWORKING, VOL. 24, NO. 3, JUNE 2016
Author: Alex X. Liu, Chad R. Meiners, and Eric Torng
Date: 2016/11/23
Department of Computer Science and Information Engineering
National Cheng Kung University, Taiwan R.O.C.
CSIE CIAL Lab 1

2. 2019/11/2
Introduction - TCAM
NOR-type TCAM cell : four comparison transistors and two SRAM cells that have a total twelve transistors.
NAND-type TCAM cell : four comparison transistors and two SRAM cells that have a total twelve transistors.
TCAM power consumption is proportional to the number of bits searched.
A TCAM chip consists of many TCAM core cells and other supporting circuits and hardware.
National Cheng Kung University CSIE Computer & Internet Architecture Lab
CSIE CIAL Lab

3. Introduction - TCAM Expensive Limited Capacity Large Power Consumption
2019/11/2
Introduction - TCAM
Expensive
Limited Capacity
Large Power Consumption
The largest available TCAM chip has a capacity of 72Mb, while 2Mb and 1Mb chips are the most popular.
Src dst port range 展開存入tcam很花空間
National Cheng Kung University CSIE Computer & Internet Architecture Lab
CSIE CIAL Lab

4. Introduction - BCAM Every bit has only two possible states: 0 or 1.
2019/11/2
Introduction - BCAM
Every bit has only two possible states: 0 or 1.
NOR-type BCAM cell : four comparison transistors and an SRAM cell.
NAND-type BCAM cell : three comparison transistors and an SRAM cell.
Power Consumption lower than TCAM.
This implies a TCAM chip consumes roughly twice
as much power and roughly twice as much board space as an
equivalent BCAM chip. Considering the number of transistors
and the associated circuits, the circuitry of a TCAM chip is
about two times denser than that of a BCAM chip.
National Cheng Kung University CSIE Computer & Internet Architecture Lab
CSIE CIAL Lab

5. B-CLASS One-dimensional packet classification using BCAM.
2019/11/2
B-CLASS
One-dimensional packet classification using BCAM.
BCAM can only store binary bits(0, 1).
Convert the ternary matching problem into a binary string exact matching problem.
National Cheng Kung University CSIE Computer & Internet Architecture Lab
CSIE CIAL Lab

6. B-CLASS - Prefix Membership Verification
2019/11/2
B-CLASS - Prefix Membership Verification
National Cheng Kung University CSIE Computer & Internet Architecture Lab
CSIE CIAL Lab

7. B-CLASS - Prefix Membership Verification
prefix numericalization
First replace every * by 0.
Second, we append bits whose value is equal to k.
National Cheng Kung University CSIE Computer & Internet Architecture Lab

8. B-CLASS - Prefix Membership Verification
2019/11/2
B-CLASS - Prefix Membership Verification
We use N(PF(B)) to denote the resulting set of binary numbers after numericalizing every prefix in PF(B). For example, N(PF(0101)) = { , , , , }.
0101, 010*, 01**, 0***, ****
National Cheng Kung University CSIE Computer & Internet Architecture Lab
CSIE CIAL Lab

9. B-CLASS
Our 1-dimensional scheme B-CLASS is composed of two algorithms.
The first algorithm describes how to preprocess a classifier so that it can be stored in BCAM.
The second algorithm classifies a packet.
National Cheng Kung University CSIE Computer & Internet Architecture Lab

10. B-CLASS - Classifier preprocessing
2019/11/2
B-CLASS - Classifier preprocessing
First convert classifier to an equivalent minimum prefix classifier C’.
Second, for each prefix P in C, we convert P to a binary number using the prefix numericalization described above. The preprocessing result N(C) is stored in BCAM.
National Cheng Kung University CSIE Computer & Internet Architecture Lab
CSIE CIAL Lab

11. B-CLASS - Classify packets
2019/11/2
B-CLASS - Classify packets
First, generate the prefixes in PF(B) in the decreasing order, We use OPF(B) to denote this ordered prefix family.
Second, numericalize every prefix in OPF(B) and get N(OPF(B)).
National Cheng Kung University CSIE Computer & Internet Architecture Lab
CSIE CIAL Lab

12. 2019/11/2
B-CLASS
National Cheng Kung University CSIE Computer & Internet Architecture Lab
CSIE CIAL Lab

13. 2019/11/2
B-CLASS - Search
The search process starts by testing whether the first element of N(OPF(B)) (i.e., N(b1b2 · · · bw)) is in the BCAM. If yes, then return the corresponding decision; otherwise, continue to test whether the second element of N(OPF(B)) is in the BCAM.
National Cheng Kung University CSIE Computer & Internet Architecture Lab
CSIE CIAL Lab

14. 2019/11/2
B-CLASS
National Cheng Kung University CSIE Computer & Internet Architecture Lab
CSIE CIAL Lab

15. B-CLASS Optimization – Skip List
2019/11/2
B-CLASS Optimization – Skip List
Given a 1-dimensional minimum prefix classifier C, we first sort all rules in C in the decreasing order of their prefix length to get Cs.
Second, we build a skip list for Cs, which is an ordered list of all the distinct lengths of the prefixes in Cs in decreasing order.
Reduce the number of BCAM lookups
For example, the skip list for the sorted minimum prefix classifier Cs in Figure 1 is 3, 2, 0. Given a packet B, we only need to search N(OPF(B))[w −i] starting from the largest number i in the skip list of Cs until a match is found. Consider the example in Figure 1, we start the search from N(OPF(B))[3] = rather than from N(OPF(B))[4] =
National Cheng Kung University CSIE Computer & Internet Architecture Lab
CSIE CIAL Lab

16. B-CLASS Optimization – Free Expansion
2019/11/2
B-CLASS Optimization – Free Expansion
Replace the rule P 11∗∗→a by two rules 110∗ → a and 111∗ → a, for all packets that match P, the number of BCAM lookups is reduced from i + 1 to i.
National Cheng Kung University CSIE Computer & Internet Architecture Lab
CSIE CIAL Lab

17. MULTI-DIMENSIONAL PACKET CLASSIFICATION
2019/11/2
MULTI-DIMENSIONAL PACKET CLASSIFICATION
Decomposing a d-dimensional packet classifier into d one dimensional classifiers.
We create the decision tree structure by first converting the given classifier to an equivalent reduced firewall decision diagram (FDD).
acyclic and directed graph
National Cheng Kung University CSIE Computer & Internet Architecture Lab
CSIE CIAL Lab

18. MULTI-DIMENSIONAL PACKET CLASSIFICATION
2019/11/2
MULTI-DIMENSIONAL PACKET CLASSIFICATION
National Cheng Kung University CSIE Computer & Internet Architecture Lab
CSIE CIAL Lab

19. MULTI-DIMENSIONAL PACKET CLASSIFICATION
2019/11/2
MULTI-DIMENSIONAL PACKET CLASSIFICATION
National Cheng Kung University CSIE Computer & Internet Architecture Lab
CSIE CIAL Lab

20. B-CLASS-d First, we numericalize every entry in the table.
2019/11/2
B-CLASS-d
First, we numericalize every entry in the table.
Second, for every entry in the table, we store the index of the corresponding skip list of the one-dimensional table that the entry belongs to.(Stored in the SRAM associated with the BCAM)
National Cheng Kung University CSIE Computer & Internet Architecture Lab
CSIE CIAL Lab

21. MULTI-DIMENSIONAL PACKET CLASSIFICATION
Search - a packet (0011, 0010), the first search key is , which returns 01. The second search key is , which returns d (i.e., discard) as the decision for the packet.
National Cheng Kung University CSIE Computer & Internet Architecture Lab

22. Lookup time
For example, the lookup time for packet (0011, 0010) on the multi-lookup table in Figure 2(c) is 5 because the first field 0011 needs 2 BCAM lookups on table 00 in 2(b) and the second field 0010 needs 3 BCAM lookups on table 10 in 2(b).
National Cheng Kung University CSIE Computer & Internet Architecture Lab

23. Algorithm for Minimizing Maximum Packet Lookup Time
2019/11/2
Algorithm for Minimizing Maximum Packet Lookup Time
Find the correct prefixes to expand such that the maximum lookup time can be reduced.
For example, the FDD in Figure 2(b) has 3 lookup paths with maximum lookup time of 6: (∗∗∗∗/00, ∗∗∗∗/01), (∗∗∗∗/00, ∗∗∗∗/10), (∗∗∗∗/00, ∗∗∗∗/11).
expanding ∗∗∗∗/00 => reduce the lookup time of all three paths.
National Cheng Kung University CSIE Computer & Internet Architecture Lab
CSIE CIAL Lab

24. Algorithm for Minimizing Maximum Packet Lookup Time
If we want to reduce the lookup time of the packets whose first matching prefix is ∗ ∗ ∗∗, we can either expand ∗ ∗ ∗∗ to four prefixes 00 ∗ ∗, 01 ∗ ∗, 10 ∗ ∗, and 11 ∗ ∗, or we can expand 00 ∗ ∗ to four prefixes 0000, 0001, 0010, and 0011.
National Cheng Kung University CSIE Computer & Internet Architecture Lab

25. Algorithm for Minimizing Average Packet Lookup Time
2019/11/2
Algorithm for Minimizing Average Packet Lookup Time
Expanding any prefix has the benefit of reducing the lookup time for some packets.
The number of packets whose lookup time will be reduced by one / the cost of expanding all prefixes of length(BCAM entry required)
Repeats until the average lookup time reduces very slowly
expanding any prefix has the benefit of reducing the lookup time for some packets and at the same time has a certain cost in terms of addition BCAM entries required.
We predefine a small threshold ε. When |AVGt−AVGt−1| AVGt−1 < ε, we terminate the process after round t. In our experiments, we chose ε = 10−2.
National Cheng Kung University CSIE Computer & Internet Architecture Lab
CSIE CIAL Lab

26. Lookup Short Circuiting
2019/11/2
Lookup Short Circuiting
Some paths may omit unnecessary fields when a node in the path contains only one outgoing edge. In this case, the node along with singleton outgoing edge can be pruned.
This optimization allows some specific decision paths to be performed with a reduced number of lookups
National Cheng Kung University CSIE Computer & Internet Architecture Lab
CSIE CIAL Lab

27. 2019/11/2
EXPERIMENTAL RESULTS
combination of skip lists, free expansion, and short circuiting techniques “light optimization”
We use max-optimization to denote the technique of minimizing the maximum lookup time in addition to light optimization,
avg-optimization to denote the technique of minimizing the average lookup time in addition to light optimization
25種CLASSIFIER
Note that without any optimizations, B-CLASS-d requires 109 lookups for every packet.
National Cheng Kung University CSIE Computer & Internet Architecture Lab
CSIE CIAL Lab

28. 2019/11/2
EXPERIMENTAL RESULTS
TCAM scheme (with short circuiting)
National Cheng Kung University CSIE Computer & Internet Architecture Lab
CSIE CIAL Lab

29. 2019/11/2
EXPERIMENTAL RESULTS
S. Dharmapurikar, P. Krishnamurthy, and D. E. Taylor, “Longest prefix
matching using bloom filters,” in Proc. ACM SIGCOMM, 2003,
pp. 201–212.
National Cheng Kung University CSIE Computer & Internet Architecture Lab
CSIE CIAL Lab