1 Kittikul Kovitanggoon*, Burin Asavapibhop, Narumon Suwonjandee, Gurpreet Singh Chulalongkorn University, Thailand July 23, 2015 Workshop on e-Science and High Performance Computing (eHPC2015) Big data management at CMS collaboration with worldwide LHC computing grid

2 2 Outline

3 3 Introductions CMS has had a distributed computing model motivated by various factors. The large quantity of data and computing requirement encouraged distributed resources from a facility infrastructure point of view. Ability to leverage resources at labs and university. Hardware, expertise, infrastructure. Benefits of providing local control of some resources. Ability to secure local funding sources. ~20% of the resources are located at CERN, 40% at T1s, and 40% at T2s, Relies on the development of tools to make transparent access to the resources. Efficient distributed computing services. Can only be successful with sufficient networking between facilities. Availability of high performance networks has made the distributed model feasible.

4 4 Large Hadron Collider (LHC) 27 km in circumference To collide rotating beams of protons or heavy ions Maximum energy of proton- proton collisions at = 14 TeV and 4 x cm -2 s -1 In 2011, collision at = 7 TeV and 4 x cm -2 s -1 In 2012, collision at = 8 TeV and 7.7 x cm -2 s -1 In 2015, expect collision at = 13 TeV and 22.8 x cm - 2 s -1 CMS ALICE ATLAS LHCb

5 5 Compact Muon Solenoid (CMS)

6 6 CMS Collisions

7 7 CMS Collision Data ● CMS detectors are gigantic digital cameras that can identify various elementary particles from the millions of collisions per second. ● Decay particles from each collisions will be: Recorded the passage of each particle through various sub-detectors as a series of electronic signals. Sent the data to the CERN Data Centre (DC) for digital reconstruction. Reconstructed digitized summary as a `collision event’. Data from the CMS experiments will be distributed around the globe with the Worldwide LHC Computing Grid (WLCG) project that is built and maintained for data storage and provides analysis infrastructure for the entire CMS community Thailand Involved in WLCG: Tier-2/Tier-3 computing centres of CMS Thailand [T2_TH_CUNSTDA and T3_TH_CHULA].

8 8 CMS Physics and Event Rates Design Luminosity (L) = cm -2 s pp events/25 ns xing ~ 1 GHz input rate “Good” events contain ~ 20 bkg. Events 1 kHz W events 10 Hz top events < 104 detectable Higgs decays/year Can store ~ 300 Hz events Select in stages Level-1 Triggers 1 GHz to 100 kHz High Level Triggers 100 kHz to 300 Hz

9 9 CMS Physics and Event Rates

10 CMS Data Flow

11 Worldwide LHC Computing Grid

12 Tier-0 The first tier in the CMS model, for which there is only one site, CERN, is known as Tier-0 (T0). The T0 performs several functions. The standard workflow is as follows: 1.accepts RAW data from the CMS Online Data Acquisition and Trigger System (TriDAS) 2.repacks the RAW data received from the DAQ into primary datasets based on trigger information 3.archives the repacked RAW data to tape 4.distributes RAW data sets among the next tier stage resources (Tier-1) so that two copies are saved 5.performs PromptCalibration in order to get the calibration constants needed to run the reconstruction 6.feeds the RAW datasets to reconstruction 7.performs prompt first pass reconstruction which writes the RECO and Analysis Object Data (AOD) extraction 8.distributes the RECO datasets among Tier-1 centers, such that the RAW and RECO match up at each Tier-1 9.distributes full AOD to all Tier-1 centers The T0 does not provide analysis resources and only operates scheduled activities.

13 Tier-1 There is a set of thirteen Tier-1 (T1) sites, which are large centers in CMS collaborating countries (large national labs, e.g. FNAL, and RAL). Tier-1 sites will in general be used for large-scale, centrally organized activities and can provide data to and receive data from all Tier-2 sites. Each T1 center: 1.receives a subset of the data from the T0 related to the size of the pledged resources in the WLCG MOU 2.provides tape archive of part of the RAW data (secure second copy) which it receives as a subset of the datasets from the T0 3.provides substantial CPU power for scheduled: re-reconstruction, skimming, calibration, and AOD extraction 4.stores an entire copy of the AOD 5.distributes RECOs, skims and AOD to the other T1 centers and CERN as well as the associated group of T2 centers 6.provides secure storage and redistribution for MC events generated by the T2's

14 Tier-2 A more numerous set of smaller Tier-2 (T2) centers but with substantial CPU resources. T2 provide: 1.services for local communities 2.grid-based analysis for the whole experiment (Tier-2 resources available to whole experiment through the grid) 3.Monte Carlo simulation for the whole experiment T2 centers rely upon T1s for access to large datasets and for secure storage of the new data (generally Monte Carlo) produced at the T2. The MC production in Tier-2's will in general be centrally organized, with generated MC samples being sent to an associated Tier-1 site for distribution among the CMS community. All other Tier-2 activities will be user driven, with data placed to match resources and needs: tape, disk, manpower, and the needs of local communities. The Tier-2 activities will be organized by the Tier-2 authorities in collaboration with physics groups, regional associations and local communities.

15 Thailand at CERN WLCG Tier-2/Tier-3 computing centres of CMS Thailand [T2_TH_CUNSTDA and T3_TH_CHULA] CMS Chulalongkorn University, Bangkok ALICE King Mongkut's University of Technology Thonburi (KMUTT), Bangkok Thai Microelectronics Center (TMEC), Muang Chachoengsao Suranaree University of Technology, Nakhon Ratchasima CERN and Thailand: relations/nms/thailand.html

16 Tier-2/Tier-3 Monitoring Monitoring Tier-2/Tier-3 of CMS Thailand that are part of worldwide LHC computing grid (WLCG). Duties for T2 TH CUNSTDA and T3 TH CHULA Periodical investigation of CMS site readiness, availability and reliability status. To maintain adequate network bandwidth and high PhEDEx – CMS data transfers. Contribute to preserve memory for analysis operations, data operations, and local activities. WLCG squid monitoring for server traffic volume, HTTP hits/requests, cached objects.

17 Tier-2/Tier-3 Monitoring

18 Tier-2/Tier-3 Monitoring

19 Tier-3 Maintenance Working with IBM system x3755 M3, x3550 M4, BladeCenter H Ensuring suitable working condition of Tier-3 system, user machines and web-server hosts, etc. Management of user accounts and corresponding data. Serving e-science etc. etc.

20 Conclusions

21 Acknowledgments This research is supported by Rachadapisek Sompote Fund for Postdoctoral Fellowship, Chulalongkorn University. Department of Physics, Faculty of Science, Chulalongkorn University for financial support. The CMS collaboration. All eHPC 2015 staffs for organizing this event.