CMS Offline and Computing Public Results

The plots show estimates from 2021 made for the CMS contribution to the LHCC November review of HL-LHC computing and common software, which supersede previous results from CMS. Coherently with the CMS Phase-2 Upgrade HLT TDR, an integrated luminosity of 270 fb-1 per year at 140 pileup events (5 KHz high level trigger rate) and a 1.2 million seconds heavy-ions run is considered during Run 4. For what concerns Run 5, an integrated luminosity of 350 fb-1 per year at 200 pileup events (7.5 KHz high level trigger rate) is considered during Run 5. Two scenarios are considered: the first one is a baseline, which does not include any improvement due to ongoing R&D activity, and the second one incorporates the most probable outcome of the ongoing R&D activities. The blue curves (and points) show the annual projected needs, summed across Tier-0, Tier-1 and Tier-2 resource needs in each of these scenarios. The gray band shows the projected resource availability for an example scenario that extrapolates the 2018 CMS pledged resources using an annual increase in available resources of between 10% and 20%. Results are derived from a bottoms-up model of CMS offline processing activities, including prompt reconstruction, Monte Carlo simulation, data re-reconstruction and all phases of analysis activities.

CPU time, disk and tape time projected requirements estimated to be required annually for CMS processing and analysis needs.


png-file pdf-file source-file

png-file pdf-file source-file

png-file pdf-file source-file

Approximate breakdown of CPU time, disk and tape requirements into primary processing and analysis activities during a typical HL-LHC year. The plots show estimates from 2021 made for the CMS contribution to the LHCC November review of HL-LHC computing and common software, which supersede previous results from CMS. The plot corresponds to a snapshot of the year 2029. The baseline scenario is considered, i.e. projected effects from on-going R&D that will reduce the computing resources needed by CMS are not considered.


png-file pdf-file

png-file pdf-file

png-file pdf-file

CPU time requirements (in kilo-HEPSpec06 years) estimated to be required annually for CMS processing and analysis needs. The plot shows estimates from 2020 made for the CMS contribution to the LHCC May review of HL-LHC computing, which supersede previous results from CMS. Two scenarios are considered: (solid line) one historically considered by CMS which assumes a running scenario reaching 275 fb-1 during production running years of Run 4, with a baseline of 7.5 kHz of data saved to the offline; and (dashed line) a common scenario for the review which assumes a running scenario of 500fb-1 per year during Run 4, with 10 kHz of data savers to the offline. The blue curves (and points) show the annual projected CPU need, summed across Tier-0, Tier-1 and Tier-2 resource needs in each of these scenarios. The black curve shows the projected resource availability for an example scenario that extrapolates the current CMS processing resources using an annual increase in available resources of between 10% and 20%. Results are derived from a bottoms-up model of CMS offline processing activities, including prompt reconstruction, Monte Carlo simulation, data re-reconstruction and all phases of analysis activities. Projected effects from on-going R&D that will reduce the computing resources needed by CMS are not included in these results.

png-file pdf-file
Disk space requirements (in PBs) estimated to be required annually for CMS processing and analysis needs. The plot shows estimates from 2020 made for the CMS contribution to the LHCC May review of HL-LHC computing, which supersede previous results from CMS. Two scenarios are considered: (solid line) one historically considered by CMS which assumes a running scenario reaching 275 fb-1 during production running years of Run 4, with a baseline of 7.5 kHz of data saved to the offline; and (dashed line) a common scenario for the review which assumes a running scenario of 500fb-1 per year during Run 4, with 10 kHz of data savers to the offline. The blue curves (and points) show the annual projected CPU need, summed across Tier-0, Tier-1 and Tier-2 resource needs in each of these scenarios. The black curve shows the projected resource availability for an example scenario that extrapolates the current CMS processing resources using an annaul increase in available resources of between 10% and 15%. Results are derived from a bottoms-up model of CMS offline processing activities, including prompt reconstruction, Monte Carlo simulation, data re-reconstruction and all phases of analysis activities. Projected effects from on-going R&D that will reduce the computing resources needed by CMS are not included in these results.

png-file pdf-file
Tape storage requirements (in PBs) estimated to be required annually for CMS processing and analysis needs. The plot shows estimates from 2020 made for the CMS contribution to the LHCC May review of HL-LHC computing, which supersede previous results from CMS. Two scenarios are considered: (solid line) one historically considered by CMS which assumes a running scenario reaching 275 fb-1 during production running years of Run 4, with a baseline of 7.5 kHz of data saved to the offline; and (dashed line) a common scenario for the review which assumes a running scenario of 500fb-1 per year during Run 4, with 10 kHz of data savers to the offline. The blue curves (and points) show the annual projected CPU need, summed across Tier-0, Tier-1 and Tier-2 resource needs in each of these scenarios. The black curve shows the projected resource availability for an example scenario that extrapolates the current CMS processing resources using an annual increase in available resources of between 10% and 15%. Results are derived from a bottoms-up model of CMS offline processing activities, including prompt reconstruction, Monte Carlo simulation, data re-reconstruction and all phases of analysis activities. Projected effects from on-going R&D that will reduce the computing resources needed by CMS are not included in these results.

png-file pdf-file
Approximate breakdown of CPU time requirements into primary processing and analysis activities during a typical HL-LHC year. The plot shows estimates from 2020 made for the CMS contribution to the LHCC May review of HL-LHC computing, which supersede previous results from CMS. The plot corresponds to the the scenario historically considered by CMS which assumes a running scenario reaching 275 fb-1 during production running years of Run 4, with a baseline of 7.5 kHz of data saved to the offline. Results are derived from a bottoms-up model of CMS offline processing activities, including prompt reconstruction, Monte Carlo simulation, data re-reconstruction and all phases of analysis activities. Projected effects from on-going R&D that will reduce the computing resources needed by CMS are not included in these results.

png-file pdf-file
Approximate breakdown of tape storage requirements into primary processing and analysis activities during a typical HL-LHC year. The plot shows estimates from 2020 made for the CMS contribution to the LHCC May review of HL-LHC computing, which supersede previous results from CMS. The plot corresponds to the the scenario historically considered by CMS which assumes a running scenario reaching 275 fb-1 during production running years of Run 4, with a baseline of 7.5 kHz of data saved to the offline. Results are derived from a bottoms-up model of CMS offline processing activities, including prompt reconstruction, Monte Carlo simulation, data re-reconstruction and all phases of analysis activities. Projected effects from on-going R&D that will reduce the computing resources needed by CMS are not included in these results.

png-file pdf-file

CPU time requirements (in kilo-HEPSpec06 years) estimated to be required annually for CMS processing and analysis needs. The plot shows estimates from 2018, which supersede previous results from CMS. The graphic has been reformatted in 2020 include an extrapolation of available resources. The blue curve (and points) show the annual projected CPU need, summed across Tier-0, Tier-1 and Tier-2 resource needs. The black curve shows the projected resource availability for an example scenario that extrapolates the current CMS processing resources using an annual increase in purchasing power of 20%. Results are derived from a bottoms-up model of CMS offline processing activities, including prompt reconstruction, Monte Carlo simulation, data re-reconstruction and all phases of analysis activities. img325d743a27d91a8b0e16d0f5a54154d3.png
Disk storage requirements (in PB) estimated to be required annually for CMS processing and analysis needs. The plot shows estimates from 2018, which supersede previous results from CMS. The graphic has been reformatted in 2020 include an extrapolation of available resources. The blue curve (and points) show the annual projected disk storage need, summed across Tier-0, Tier-1 and Tier-2 resource needs. The black curve shows the projected resource availability for an example scenario that extrapolates the current CMS processing resources using an annual increase in purchasing power of 15%. Results are derived from a bottoms-up model of CMS offline processing activities, including prompt reconstruction, Monte Carlo simulation, data re-reconstruction and all phases of analysis activities. img3ab31c662352f2c5ff15a9d79c967313.png
Tape storage requirements (in PB) estimated to be required annually for CMS processing and analysis needs. The plot shows estimates from 2018, which supersede previous results from CMS. The graphic has been reformatted in 2020 include an extrapolation of available resources. The blue curve (and points) show the annual projected tape storage need, summed across Tier-0 and Tier-1 resource needs.. The black curve shows the projected resource availability for an example scenario that extrapolates the current CMS processing resources using an annual increase in purchasing power of 15%. Results are derived from a bottoms-up model of CMS offline processing activities, including prompt reconstruction, Monte Carlo simulation, data re-reconstruction and all phases of analysis activities. img874aeb2e38618e41666aae931c2e94d2.png
CPU time requirements (in Tera-HEPSpec06 seconds) estimated to be required annually for CMS processing and analysis needs. The plot shows estimates from 2018, which supersede previous results from CMS. The annual bars show the annual projected CPU need, summed across Tier-0, Tier-1 and Tier-2 resource needs. As denoted in the legend, the total resource need is separated into its largest components. Results are derived from a bottoms-up model of CMS offline processing activities, including prompt reconstruction, Monte Carlo simulation, data re-reconstruction and all phases of analysis activities. img8c53bf7c4d502234891b5e8f1ab84724.png
Disk storage requirements (in PB) estimated to be required annually for CMS processing and analysis needs. The plot shows estimates from 2018, which supersede previous results from CMS. The annual bars show the annual projected disk need, summed across Tier-0, Tier-1 and Tier-2 resource needs. As denoted in the legend, the total resource need is separated into its largest components. Results are derived from a bottoms-up model of CMS offline processing activities, including prompt reconstruction, Monte Carlo simulation, data re-reconstruction and all phases of analysis activities. img2b7f17763c6ea9165d3644f2f3b5f7ed.png
Tape storage requirements (in PB) estimated to be required annually for CMS processing and analysis needs. The plot shows estimates from 2018, which supersede previous results from CMS. The annual bars show the annual projected disk need, summed across Tier-0 and Tier-1 resource needs. As denoted in the legend, the total resource need is separated into its largest components. Results are derived from a bottoms-up model of CMS offline processing activities, including prompt reconstruction, Monte Carlo simulation, data re-reconstruction and all phases of analysis activities. imgbc5303a312965c52e9275fd1de1a1b8a.png
Edit | Attach | Watch | Print version | History: r15 < r14 < r13 < r12 < r11 | Backlinks | Raw View | WYSIWYG | More topic actions
Topic revision: r15 - 2022-01-10 - DaniloPiparo
 
    • Cern Search Icon Cern Search
    • TWiki Search Icon TWiki Search
    • Google Search Icon Google Search

    CMSPublic All webs login

This site is powered by the TWiki collaboration platform Powered by PerlCopyright & 2008-2022 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
or Ideas, requests, problems regarding TWiki? use Discourse or Send feedback