ABSTRACT
Reducing the power consumption of core Internet routers is important for both Internet Service Providers (ISPs) and router vendors. ISPs can reduce their Carbon footprint and operational costs, while router manufacturers can achieve higher switching capacity per rack space. In this work, we examine the impact of packet buffers on the power consumption of backbone router line-cards. We argue that Gigabytes of always-on SRAM and DRAM buffers account for around 10% of the power, but are actively used only during transient periods of congestion. We propose a simple and practical algorithm for activating buffers incrementally as needed and putting them to sleep when not in use. We evaluate our algorithm on traffic traces from carrier and enterprise networks, via simulations in ns2, and by implementing it on a programmable-router test-bed. Our study shows that much of the energy associated with off-chip packet buffers can be eliminated with negligible impact on traffic performance. Dynamic adjustment of active router buffer size provides a low-complexity low-risk mechanism of saving energy that is amenable for incremental deployment in networks today.
- M. Pickavet. Network Solutions to Reduce the Energy Footprint of ICT. In European Conference on Optical Communication (ECOC) Symposium, Belgium, Sep 2008.Google Scholar
- R. Tucker et al. Energy Consumption of IP Networks. In Proc. Europ. Conf. on Optical Comm., Belgium, Sep 2008.Google ScholarCross Ref
- GreenTouch. www.greentouch.org.Google Scholar
- S. Iyer, R. R. Kompella, and N. McKeown. Designing packet buffer for router linecards. IEEE/ACM Trans. on Networking, 16(3):705--717, Jun 2008. Google ScholarDigital Library
- Micron-Technology-Inc. System power Calculators. www.micron.com/support/dram/power_calc.html.Google Scholar
- Cypress-Semiconductor. QDR-II SRAM CY7C1515KV18. http://www.cypress.com/?docID=24145.Google Scholar
- R. Tucker et al. Evolution of WDM Optical IP Networks: A Cost and Energy Perspective. IEEE/OSA J. of Lightwave Tech., 27(3):243--252, Feb 2009.Google ScholarCross Ref
- S. Aleksic. Analysis of Power Consumption in Future High-Capacity Network Nodes. IEEE/OSA J. Optical Comm. and Netw., 1(3):245--258, Aug 2009.Google ScholarCross Ref
- O Tamm. Scaling and Energy Efficiency in Next Generation Core Networks and Switches. In ECOC Sunday Workshop., Austria, Sep 2009.Google Scholar
- G. Epps et al. System Power Challenges. Cisco Research Seminar, Aug 2006. http://www.cisco.com/web/about/ac50/ac207/proceedings/POWER_GEPPS_rev3.ppt.Google Scholar
- AMD Opteron processors. http://www.amd.com/acp.Google Scholar
- EZchip 50 Gbps NP-4 Network Processor. http://www.ezchip.com/Images/pdf/NP-4_Short_Brief_online.pdfGoogle Scholar
- Internet2 Network. http://www.internet2.edu/network/.Google Scholar
- G. Shen and R. Tucker. Energy-Minimized Design for IP over WDM Networks. IEEE/OSA J. of Optical Comm. and Netw., 1(1):176--186, Jun 2009.Google ScholarCross Ref
- J. Chabarek et al. Power Awareness in Network Design and Routing. In Proc. IEEE INFOCOM, USA, Mar 2008.Google ScholarCross Ref
- S. Nedevschi et al. Reducing Network Energy Consumption via Rate-Adaptation and Sleeping. In Proc. USENIX NSDI, USA, Apr 2008. Google ScholarDigital Library
- Bill St.Arnaud. CANARIE: Research Networks to Help Reduce Global Warming. In OFC Workshop on Energy Footprint of ICT: Forecast and Netw. Sol., USA, Mar 2009.Google Scholar
- A. Cianfrani et al. An Energy Saving Routing Algorithm for a Green OSPF Protocol. In IEEE INFOCOM Traffic Monitoring and Management Workshop, USA, Mar 2010.Google ScholarCross Ref
- Y. Zhang, P. Chowdhury, M. Tornatore, and B. Mukherjee. Energy Efficiency in Telecom Optical Networks. IEEE Comm. Surveys and Tutorials, 12(4):441--458, Q4 2010. Google ScholarDigital Library
- G. Appenzeller, I. Keslassy, and N. McKeown. Sizing router buffers. In Proc. ACM SIGCOMM, USA, Aug-Sep 2004. Google ScholarDigital Library
- M. Enachescu et al. Routers with very small buffers. In Proc. IEEE INFOCOM, Spain, Apr 2006.Google ScholarCross Ref
- A. Vishwanath, V. Sivaraman, and M. Thottan. Perspectives on Router Buffer Sizing: Recent Results and Open Problems. ACM CCR, 39(2):34--39, Apr 2009. Google ScholarDigital Library
- Y. Zhang and D. Loguinov. ABS: Adaptive Buffer Sizing for Heterogeneous Networks. Computer Networks, 54(14):2562--2574, Oct 2010. Google ScholarDigital Library
- X. Wu et al. Power and Performance of Read-Write Aware Hybrid Caches with Non-Volatile Memories. In Design, Automation and Test in Europe, France, Apr 2009. Google ScholarDigital Library
- I. Norros. On the use of Fractional Brownian Motion in the Theory of Connectionless Traffic. IEEE J. Selected Areas in Comm., 13(6):953--962, Aug 1995. Google ScholarDigital Library
- D. Ostry. Synthesis of Accurate Fractional Gaussian Noise by Filtering. IEEE Trans. Information Therory, 52(4):1609--1623, Apr 2006. Google ScholarDigital Library
- CAIDA packet length distributions. http://www.caida.org/analysis/AIX/plen\_hist/.Google Scholar
- Voxel SLAs. https://www.voxel.net/sla.Google Scholar
- NetFPGA 1G and 10G Routers. www.netfpga.org.Google Scholar
- Ixia traffic generator. www.ixiacom.com.Google Scholar
- Anue Systems Network Emulator. www.anuesystems.com.Google Scholar
Recommendations
Sizing router buffers
All Internet routers contain buffers to hold packets during times of congestion. Today, the size of the buffers is determined by the dynamics of TCP's congestion control algorithm. In particular, the goal is to make sure that when a link is congested, ...
Sizing router buffers
SIGCOMM '04: Proceedings of the 2004 conference on Applications, technologies, architectures, and protocols for computer communicationsAll Internet routers contain buffers to hold packets during times of congestion. Today, the size of the buffers is determined by the dynamics of TCP's congestion control algorithm. In particular, the goal is to make sure that when a link is congested, ...
Comments