![Page 1: Power Virtualization in Multi-tenant Networks€¦ · Networking In datacenter, ~15% Flat-fee, or byte-based Power oblivious Least energy proportional; Idle power = ~80% peak; 2 power](https://reader033.vdocuments.mx/reader033/viewer/2022060605/605abd99e0576810e1025aa7/html5/thumbnails/1.jpg)
Power Virtualization in Multi-tenant Networks
Srini SeetharamanDeutsche Telekom R&D Lab USA
Clean Slate Lab, Stanford UniversityOct 2010
![Page 2: Power Virtualization in Multi-tenant Networks€¦ · Networking In datacenter, ~15% Flat-fee, or byte-based Power oblivious Least energy proportional; Idle power = ~80% peak; 2 power](https://reader033.vdocuments.mx/reader033/viewer/2022060605/605abd99e0576810e1025aa7/html5/thumbnails/2.jpg)
Deutsche Telekom Inc. R&D Lab USA
Consumption
Networking
In datacenter, ~15%
Flat-fee, or byte-based
Power oblivious
Least energy proportional;Idle power = ~80% peak;
2 power modes: on/off
Computing
In datacenter, ~45%
Billed by hourly rate of use
Power aware
More energy proportional; Idle power = ~30% peak
Multiple power modes
![Page 3: Power Virtualization in Multi-tenant Networks€¦ · Networking In datacenter, ~15% Flat-fee, or byte-based Power oblivious Least energy proportional; Idle power = ~80% peak; 2 power](https://reader033.vdocuments.mx/reader033/viewer/2022060605/605abd99e0576810e1025aa7/html5/thumbnails/3.jpg)
Deutsche Telekom Inc. R&D Lab USA
ElasticTree
• Based on a given workload, pack flows into fewer devices and turn-off unused elements
Core
Aggregation
EdgePod 0 Pod 1 Pod 2 Pod 3
![Page 4: Power Virtualization in Multi-tenant Networks€¦ · Networking In datacenter, ~15% Flat-fee, or byte-based Power oblivious Least energy proportional; Idle power = ~80% peak; 2 power](https://reader033.vdocuments.mx/reader033/viewer/2022060605/605abd99e0576810e1025aa7/html5/thumbnails/4.jpg)
Workload
A
D
E FC
B
A
A
D
E F
F
A
EC
C
BD
B
EC
D
VNet1
VNet2
Tenant
Infrastructure
Tenant 2 = $X2 , Y2 joules
Tenant 1 = $X1 , Y1 joules
![Page 5: Power Virtualization in Multi-tenant Networks€¦ · Networking In datacenter, ~15% Flat-fee, or byte-based Power oblivious Least energy proportional; Idle power = ~80% peak; 2 power](https://reader033.vdocuments.mx/reader033/viewer/2022060605/605abd99e0576810e1025aa7/html5/thumbnails/5.jpg)
Deutsche Telekom Inc. R&D Lab USA
Proposal
• Can we provide incentive to align workload in a power-aware manner?– By making usage charge of tenant
proportional to its energy consumed• Virtual power• How to determine virtual power in a non-
proportional network?
![Page 6: Power Virtualization in Multi-tenant Networks€¦ · Networking In datacenter, ~15% Flat-fee, or byte-based Power oblivious Least energy proportional; Idle power = ~80% peak; 2 power](https://reader033.vdocuments.mx/reader033/viewer/2022060605/605abd99e0576810e1025aa7/html5/thumbnails/6.jpg)
Deutsche Telekom Inc. R&D Lab USA
Heuristics
Virtual_power tenant i
= Σ Virtual_power element j
= Σ Power element j
# sharing tenants
![Page 7: Power Virtualization in Multi-tenant Networks€¦ · Networking In datacenter, ~15% Flat-fee, or byte-based Power oblivious Least energy proportional; Idle power = ~80% peak; 2 power](https://reader033.vdocuments.mx/reader033/viewer/2022060605/605abd99e0576810e1025aa7/html5/thumbnails/7.jpg)
Deutsche Telekom Inc. R&D Lab USA
Consequences
• Tenant penalized for being only occupant• Encourages reuse of pre-paid / pre-
powered-on elements• One tenant unaware of other tenants• One step closer to virtualizing networks
![Page 8: Power Virtualization in Multi-tenant Networks€¦ · Networking In datacenter, ~15% Flat-fee, or byte-based Power oblivious Least energy proportional; Idle power = ~80% peak; 2 power](https://reader033.vdocuments.mx/reader033/viewer/2022060605/605abd99e0576810e1025aa7/html5/thumbnails/8.jpg)
Deutsche Telekom Inc. R&D Lab USA
Implementation
• PowerVisor acts as a metering proxy between switches and tenants– Translates true power to virtual power
PowerVisor
Tenant A Tenant B
![Page 9: Power Virtualization in Multi-tenant Networks€¦ · Networking In datacenter, ~15% Flat-fee, or byte-based Power oblivious Least energy proportional; Idle power = ~80% peak; 2 power](https://reader033.vdocuments.mx/reader033/viewer/2022060605/605abd99e0576810e1025aa7/html5/thumbnails/9.jpg)
Deutsche Telekom Inc. R&D Lab USA
Billing
Multiple ways of monetizing the energy consumed:
• Directly proportional to the energy footprint
• Auction resources to tenants for flow usage
• Finite energy allocated that depletes in a capacitor model
![Page 10: Power Virtualization in Multi-tenant Networks€¦ · Networking In datacenter, ~15% Flat-fee, or byte-based Power oblivious Least energy proportional; Idle power = ~80% peak; 2 power](https://reader033.vdocuments.mx/reader033/viewer/2022060605/605abd99e0576810e1025aa7/html5/thumbnails/10.jpg)
Deutsche Telekom Inc. R&D Lab USA
Fineprint
• Infrastructure– Conserve by powering down devices
(or choosing other available low energy states)– Fair across tenants; No cheating
• Tenant– Nothing blocks a tenant with infinite finances– Can possibly do the following:
• Load-based conflict• Collusion among tenants• Masquerading as multiple tenants• Energy trading
![Page 11: Power Virtualization in Multi-tenant Networks€¦ · Networking In datacenter, ~15% Flat-fee, or byte-based Power oblivious Least energy proportional; Idle power = ~80% peak; 2 power](https://reader033.vdocuments.mx/reader033/viewer/2022060605/605abd99e0576810e1025aa7/html5/thumbnails/11.jpg)
Deutsche Telekom Inc. R&D Lab USA
Future
• Build emulation prototype over mininet and then extend to an actual cluster– To understand the dynamics and interactions
• How can we achieve good network performance, while conserving power?
![Page 12: Power Virtualization in Multi-tenant Networks€¦ · Networking In datacenter, ~15% Flat-fee, or byte-based Power oblivious Least energy proportional; Idle power = ~80% peak; 2 power](https://reader033.vdocuments.mx/reader033/viewer/2022060605/605abd99e0576810e1025aa7/html5/thumbnails/12.jpg)
Backup
![Page 13: Power Virtualization in Multi-tenant Networks€¦ · Networking In datacenter, ~15% Flat-fee, or byte-based Power oblivious Least energy proportional; Idle power = ~80% peak; 2 power](https://reader033.vdocuments.mx/reader033/viewer/2022060605/605abd99e0576810e1025aa7/html5/thumbnails/13.jpg)
Deutsche Telekom Inc. R&D Lab USA
Previous analysis
• Energy Dumpster Diving
![Page 14: Power Virtualization in Multi-tenant Networks€¦ · Networking In datacenter, ~15% Flat-fee, or byte-based Power oblivious Least energy proportional; Idle power = ~80% peak; 2 power](https://reader033.vdocuments.mx/reader033/viewer/2022060605/605abd99e0576810e1025aa7/html5/thumbnails/14.jpg)
Deutsche Telekom Inc. R&D Lab USA
Previous analysis (contd.)
• Power Benchmarking Framework for Network Devices
![Page 15: Power Virtualization in Multi-tenant Networks€¦ · Networking In datacenter, ~15% Flat-fee, or byte-based Power oblivious Least energy proportional; Idle power = ~80% peak; 2 power](https://reader033.vdocuments.mx/reader033/viewer/2022060605/605abd99e0576810e1025aa7/html5/thumbnails/15.jpg)
Deutsche Telekom Inc. R&D Lab USA
Previous analysis (contd.)
• The cost of a cloud: Research problems in data center networks