How Green is IPTelephony Salman Abdul Baset Joshua
How Green is IP-Telephony? Salman Abdul Baset*, Joshua Reich*, Janak**, Pavel Kasparek**, Vishal Misra*, Dan Rubenstein*, Henning Schulzrinne* Department of Computer Science, Columbia University* Tekelec Corporation**
Traditional Telephony • Place call (Signaling) • Directory lookup • Circuit reservation • Talk (Connectivity) • Transfer voice data (analog, digital) • Variations on these themes • Multi-party conferencing • Voicemail 2
IP-Based Communication Systems Telephony • Place call (Signaling) • Directory lookup • Packet switched routes • Talk (Connectivity) And More • Video • IM • Status / buddy list • Direct packet routing • Media relaying • PSTN/mobile gateways 3
Trends & Implications • Trend away from traditional telephony infrastructure • Vonage, Packet 8, Verizon Fi. OS • Skype • Mobile • To single infrastructure (IP) for all data/voice/etc. • More efficient (one system to maintain, improve) • Much less expensive (for now) • More fragile (one system to fail) • More complex So what does this mean energy wise? 4
Our Questions 1) Where is energy consumed? 2) How do different design choices effect energy consumption? 3) How we can make IP-telephony more energy efficient? 5
Outline • IP-Telephony and power consumption • Answering our questions: 1) Where is energy consumed? 2) How do different design choices effect energy consumption? 3) How we can make IP-telephony more energy efficient? • Conclusion & Future Work 6
Outline • IP-Telephony and power consumption • Answering our questions: 1) Where is energy consumed? 2) How do different design choices effect energy consumption? 3) How we can make IP-telephony more energy efficient? • Conclusion & Future Work 7
IP Communication Flavors Client-Server (C/S) Peer-to-Peer (P 2 P) 8
IP Communication Flavors Traditional Telephony Replacement Communication Addendum 9
How Does C/S IP-Telephony Work? SIPSIP registrar / / registrar proxy / presence proxy server REGISTER (1) signaling (ip addr) User agent IP-PSTN gateway (2) media (voice, video, IM) User agent Utopian Internet No NATs or firewalls 10
And In The Real-World… SIP registrar / proxy / presence / server media server NAT / firewall User agent 11
Media Servers Bypass Firewalls SIP registrar / proxy / presence / server media server IP-PSTN gateway (1) signaling NAT / firewall (2) media (1) signaling NAT / firewall (voice, video, IM) (UDP or TCP) User agents 12
How Does P 2 P IP-Telephony Work? media relay (or relay) node A node E NAT / firewall network address media of(3) node B? (TCP) (2) (4) media (2) signaling (1) (3) signaling (1) node B NAT / firewall (1) network address (2) of node E? (1) (2) signaling node C node = user agent P 2 P / PSTN gateway • • node D • nodes form an overlay share responsibilities for message routing, signaling, media relaying super nodes, ordinary nodes 13
Sources of Energy Consumption • End-point – Handsets – Vo. IP conversion boxes – PCs • Core – Signaling / directory – Media relaying – PSTN / mobile gateways • Network 14
Assessing Energy Consumption • Data (from C/S Vo. IP provider) • Power Meters – – – 100 K users (mostly business) 15 calls per second (CPS) ~5 K calls in system NAT keep-alive traffic All calls relayed • Modeling – C/S – P 2 P – Wattsup – Killawatt • Hardware Measurements – – – – SIP Server Relay Server Desktop clients Laptop clients Hardware SIP phones Software phones Skype peers 15
Outline • IP-Telephony and power consumption • Answering our questions: 1) Where is energy consumed? 2) How do different design choices effect energy consumption? 3) How we can make IP-telephony more energy efficient? • Conclusion & Future Work 16
Where is Energy Consumed? PSTN replacement • Vo. IP servers consume less than 0. 04% of total! – – >10 K users, voice traffic a server can handle signaling workload for 500 k users a server can handle media workload for 50 k users even after a redundancy factor of 2, and conservative PUE of 2! Make PSTN replacement green? Reduce end-device power consumption. 17
Where is Energy Consumed? Non-PSTN replacement • More complicated • If softphone draws little additional power – Still likely that end-point biggest component – But may not dominate consumption • If users leave PCs on just as phones – Possibly even worse than PSTN! . User / hardware study needed. 18
How Do Design Choices Effect Power Consumption? SIP registrar / proxy / presence / server media relay (or relay) node A NAT / firewall (3) media (TCP) INVITE media INVITE (2) signaling NAT / firewall (voice, video, IM) (UDP or TCP) User agents node E (1) NAT / node B firewall of node E? (2) signaling User agents • C/S Inefficiencies (1) network address node C node D – Power utilization efficiency – (PUE) • Ratio of data center power draw to IT power draw • e. g. , cooling, network equipment, etc. – Idle power consumption (can be addressed in larger systems by techniques such as Somniloquy or Sleep Proxy – Percentage of user population that requires relaying major 19 determinant of core energy consumption.
How Do Design Choices Effect Power Consumption? SIP registrar / proxy / presence / server media relay (or relay) node A NAT / firewall (3) media (TCP) INVITE media INVITE • P 2 P Consumption (2) signaling NAT / firewall (voice, video, IM) (UDP or TCP) User agents node E (1) NAT / node B firewall (1) network address of node E? (2) signaling User agents node C node D – Avoids these overheads by using machines that are already on • In theory general user population • In practice appears to be heavily subsidized by university machines • What happens when machines are idling less… – Incurs small additional energy use for signaling and relaying • But how small? 20
Comparing C/S and P 2 P • Compare under same load – Active calls – Call duration – Percentage of PSTN calls • Generic C/S and P 2 P – Both use standard Vo. IP (e. g. , not Skype) • Isolate only services that differ between P 2 P, CS – – Directory service Call signaling Media session Presence 21
Modeling P 2 P and C/S • C/S model • P 2 P model – C/S power consumption = #servers * Watts/server * redundancy factor * PUE – S super nodes active – ps super node consumption P 2 P energy efficient when: S * ps < C/S power consumption • One active super node per relayed call. • Media server fully loaded. • 100% calls relayed ps = 52 m. W P 2 P may consume more than C/S! 22
Caveats • Peers – External meters do not provide sufficient resolution to determine ps w/ confidence – Will be in different states when relay starts • Medium load unlikely to incur much extra overhead • Low or high loads, ps could conceivably be large • Consequently, prior distribution effects efficiency • Servers – Energy usage not linear w/ load – Lower utilization hurts energy efficiency 23
Making IP-Telephony Greener • Make phones energy efficient – LCD, processor, WOL for phones? • NATs & Firewalls – Get rid of NATs or rearchitect them – Use TCP to avoid NAT keep-alive – Make firewalls Vo. IP-friendly. • Set up SIP user agents on gateways • PC wakeup on receiving calls 24
Outline • IP-Telephony and power consumption • Answering our questions: 1) Where is energy consumed? 2) How do different design choices effect energy consumption? 3) How we can make IP-telephony more energy efficient? • Conclusion & Future Work 25
Conclusions • Vo. IP endpoints dominate total energy consumption in PSTN replacement systems • P 2 P not necessarily more energy efficient than C/S. • NATs and firewalls create the need for media relaying, one of the biggest components of core energy consumption. 26
Future Work • Obtain data on PSTN power consumption • Work on accurately measuring ps • Measure path length / routing differences between of direct and media-relayed calls. • Study user behavior viz-a-vis softphone use – How much extra time are machines left on – Power draw during those periods • Develop WOL capable hard-phones 27
- Slides: 27