1 / 14

Data Center Specification

Data Center Specification. Gregory Provan CS6404. Global Properties. Global data structures Global clock System inputs: job specification Sub-systems Input/output computations. Plant Model. Data Center. CRAC. Key Global data structures. R 11 . . . . . . . . . . . . . . . . R 1k

ivria
Download Presentation

Data Center Specification

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Data Center Specification Gregory Provan CS6404

  2. Global Properties • Global data structures • Global clock • System inputs: job specification • Sub-systems • Input/output • computations

  3. Plant Model Data Center CRAC

  4. Key Global data structures R11 . . . . . . . . . . . . . . . . R1k R21 . . . . . . . . . . . . . . . . R2k Rm1 . . . . . . . . . . . . . . . . Rmk • CPU matrix C (m x k) • CPU health H (m x k) • H: C  h, h{0,1} • CPU thermal output O(m x k) • O: C  T, with T [0,…,Tmax] (kW) • Thermal map T (m x k) • T: C   (degrees C) • CRAC parameters • Thermal outputs: Temp x mass-flowrate

  5. System Clock • The two basic metrics • Job-epoch j • CRAC-epoch for integer k: C = kj • CRAC is constant during this epoch • System has temporal window j • Used to compute future schedules, temperatures, etc.

  6. Jobs and Allocation • Jobs arrive at rate r, and we assume jobs arrive in batches, at job-epochs j apart • J={J1,…,Jn} • Each job denoted by pair (d,), d,  integers • d denotes the job duration •  denotes the deadline, which is  j longer than the duration • Assignment A: J x   C • Can only assign jobs to CPUs that are not expected to fail during the job duration •  is the CPU processing-rate: {low, high}

  7. Job Revenue • Revenue function  • Amount based on fulfilling the deadlines, with penalties for missing deadlines • R: J x A x    (euros)

  8. Health Computation • Requirements • Health decreases with high temp and use • Health increases with low temp and no use • If hhmax, then the CPU fails • CPUs also fail randomly with rate /day • H(t+1)  H(t) x T(t) x A(t) x 

  9. Job Scheduling • Inputs • Jobs J(t) • Weight function: w(t)=(wrevenue, wthermal) • Governs schedule returned • Prior data: A(t-1), H(t-1), T(t-1) • Output • A(t),…,A(t+) •  specifies the temporal window for the system • Computation • A(t) = f(J(t),A(t-1), H(t-1), T(t-1),w(t))

  10. Power Computation • CPU power • Each CPU consumes power based on its rate  • PC:    (kW) • Total power: A x PC • CRAC power • CRAC cooling rate: based on Tr x T* x flow-rate • Return-air temp Tr • Setpoint temp T*

  11. Thermal Computation • Used history of CPU usage, CRAC output and data-center airflow to compute thermal map • For our purposes, thermal map is T • T: C   (degrees C) • Must specify • thermal maps over CRAC time-windows • Critical-temp maps, where Tcrit: C  (-Tmax) (degrees C)

  12. Hot-Spot Computation • Assume simple air-circulation model for Rack Warm air outflow Temperature Gradient (with equal Loads) Cold air inflow

  13. Hot-Spot Computation • Impact or radiation on adjacent Racks (optional) Warm air outflow Temperature Gradient (with equal Loads) Cold air inflow

  14. Key Publication • Banerjee et al. • Integrating cooling awareness with thermal aware workload placement, Sustainable Computing 1, 2011.

More Related