在軟體定義網路中，目前大部分網路交換機的路由表是由三元可定址內容記憶體(Ternary Content Addressable Memory, TCAM)所製成。雖然TCAM可以用很快的速度來處理進入交換機的封包，但是和隨機存取記憶體(Random Access Memory, RAM)相比，此類記憶體有四個缺點，記憶體空間不夠、能量消耗大、過熱、晶片面積大。規則快取(Rules caching)則是一個重要方法來解決記憶體空間不夠的問題。但規則會彼此重疊則是一大問題。在這篇論文中，我們利用覆蓋集合(Cover set)來解決規則重疊問題。和其他的規則快取演算法相比，我們會分別計算一群規則的累積貢獻值，然後將最有貢獻的那群規則放入快取。另一方面，我們的快取替換(Cache replacement)演算法會根據封包的時間和地域的局部性，動態地將封包應該配對到的規則和快取裡的規則做替換。模擬結果顯示，我們的規則快取演算法得到平均10%的改進相較於其他規則快取演算法。我們快取替換演算法的快取命中率比起最久未使用和隨機替換演算法可分別高出12%和17%。

In Software-Defined Networking, flow tables of OpenFlow switches are implemented by ternary content addressable memory (TCAM). Although TCAM can process input packets in high speed, TCAM has four shortcomings, TCAM capacity, power consumption, heat generation, and board space problems. Rules caching is a technique to solve the TCAM capacity problem which is the most important problem in TCAM. However, the rule dependency problem is a challenging issue for wildcard rules caching where packets could mismatch rules. In this paper, we use cover set to solve rule dependency problem and cache important rules to TCAM. Instead of calculating contribution value of an individual rule, our wildcard rules caching algorithm calculates contribution value of a set of rules. Besides, we propose a cache replacement algorithm considering the temporal and spatial traffic localities. Simulation results show that our caching algorithm could get 10% average improvement ratio than previous work. Cache replacement algorithm could get 12% and 17% higher cache hit ratio than least recently used and random replacement algorithms, respectively.

I. Introduction 1
II. Related Work 5
2.1. Packet Classifications Compression 5
2.2. Distributing Rules along Traffic Route 6
2.3. Rules Caching 7
III. Algorithms 10
3.1. System Model and Problem Formulation 10
3.2. Wildcard Rules Caching Algorithm 12
3.3. Cache Replacement Algorithm 23
IV. Simulation 27
4.1. Simulation Environment 27
4.2. Simulation Results 28
V. Conclusion 37
References 38

[1] M. Reitblatt, N. Foster, J. Rexford, C. Schlesinger, D. Walker, “Abstractions for Network Update,” Proceedings of ACM SIGCOMM, pp. 323-334, Helsinki, Finland, August 2012.
[2] N. P. Katta, J. Rexford, D. Walker, “Incremental Consistent Updates,” Proceedings of the Second ACM SIGCOMM Workshop on Hot Topics in Software Defined Networking, pp. 49-54, Hong Kong, China, August 2013.
[3] X. Jin, H. H. Liu, R. Gandhi, S. Kandula, R. Mahajan, M. Zhang, J. Rexford, R. Wattenhofer, “Dynamic Scheduling of Network Updates,” Proceedings of ACM SIGCOMM, pp. 539-550, Chicago, IL, USA, August 2014.
[4] N. Gude, T. Koponen, J. Pettit, B. Pfaff, M. Casado, N. McKeown, S. Shenker, “NOX: Towards an Operating System for Networks,” ACM SIGCOMM Computer Communication Review, vol. 38, no. 3, pp. 105-110, July 2008.
[5] http://www.projectfloodlight.org/floodlight/
[6] http://osrg.github.io/ryu/
[7] N. McKeown, T. Anderson, H. Balakrishnan, G. Parulkar, L. Peterson, J. Rexford, S. Shenker, J. Turner, “Openflow: Enabling Innovation in Campus Networks,” ACM SIGCOMM Computer Communication Review, vol. 38, no. 2, pp. 69-74, April 2008.
[8] K. Lakshminarayanan, A. Rangarajan, S. Venkatachary, “Algorithms for Advanced Packet Classification with Ternary CAMs,” Proceedings of ACM SIGCOMM, pp. 193-204, Philadelphia, PA, USA, August 2005.
[9] K. Kogan, S. Nikolenko, O. Rottenstreich, W. Culhane, P. Eugster, “SAX-PAC (Scalable And eXpressive PAcket Classification),” Proceedings of ACM SIGCOMM, pp. 15-26, Chicago, IL, USA, August 2014.
[10] M. Casado, M. J. Freedman, J. Pettit, J. Luo, N. Gude, N. McKeown, S. Shenker, “Rethinking Enterprise Network Control,” IEEE/ACM Trans. on Networking, vol. 17, no. 4, pp. 1270-1283, August 2009.
[11] C. R. Meiners, A. X. Liu, E. Torng, “Bit Weaving: A Non-Prefix Approach to Compressing Packet Classifiers in TCAMs,” IEEE/ACM Trans. on Networking, vol. 20, no. 2, pp. 488-500, April 2012.
[12] Yu-Chieh Cheng, Pi-Chung Wang, “Packet Classification Using Dynamically Generated Decision Trees,” IEEE Trans. on Computers, vol. 64, no. 2, pp. 582-586, February 2015.
[13] A. X. Liu, C. R. Meiners, E. Torng, “TCAM Razor: A Systematic Approach Towards Minimizing Packet Classifiers in TCAMs,” IEEE/ACM Trans. on Networking, vol. 18, no. 2, pp. 490-500, April 2010.
[14] Xuan-Nam Nguyen, D. Saucez, C. Barakat, T. Turletti, “Optimizing Rules Placement in OpenFlow Networks: Trading Routing for Better Efficiency,” Proceedings of the Third Workshop on Hot Topics in Software Defined Networking, pp. 127-132, Chicago, IL, USA, August 2014.
[15] Y. Kanizo, D. Hay, and I. Keslassy, “Palette: Distributing tables in software-defined networks,” Proceedings of INFOCOM, pp. 545-549, Turin, Italy, April 2013.
[16] N. Kang, Z. Liu, J. Rexford, D. Walker, “Optimizing the “One Big Switch” Abstraction in Software-Defined Networks,” Proceedings of the Ninth ACM Conference on Emerging Networking Experiments and Technologies, pp. 13-24, Santa Barbara, CA, USA, December 2013.
[17] B. Yan, Y. Xu, H. Xing, K. Xi, H. J. Chao, “CAB: A Reactive Wildcard Rule Caching System for Software-Defined Networks,” Proceedings of the Third Workshop on Hot Topics in Software Defined Networking, pp. 163-168, Chicago, IL, USA, August 2014.
[18] M. Yu, J. Rexford, M. J. Freedman, J. Wang, “Scalable Flow-Based Networking with DIFANE,” Proceedings of ACM SIGCOMM, pp. 351-362, New Delhi, India, August 2010.
[19] N. Katta, O. Alipourfard, J. Rexford, D. Walker, “Infinite CacheFlow in Software-Defined Networks,” Proceedings of the Third Workshop on Hot Topics in Software Defined Networking, pp. 175-180, Chicago, IL, USA, August 2014.
[20] D. E. Taylor, J. S. Turner, “ClassBench: A Packet Classification Benchmark,” IEEE/ACM Trans. on Networking, pp. 499-511, vol. 15, no. 3, June 2007.
[21] M. Kharbutli, R. Sheikh, “LACS: A Locality-Aware Cost-Sensitive Cache Replacement Algorithm,” IEEE Trans. on Computers, vol. 63, no. 8, pp. 1975-1987, August 2014.
[22] G. Borradaile, B. Heeringa, G. Wilfong, “The Knapsack Problem with Neighbour Constraints,” Journal of Discrete Algorithms, vol. 16, pp. 224-235, October 2012.