| 1 |
fisk |
1.1 |
\section{Definition}
|
| 2 |
|
|
|
| 3 |
|
|
The combined Computing, Software, and Analysis challenge of 2006 is an
|
| 4 |
|
|
O(50) million event exercise to test the workflow and dataflow
|
| 5 |
|
|
associated with the data handling model of CMS. It is designed to be a
|
| 6 |
|
|
25\% capacity test of what is needed for operations in 2008. The main
|
| 7 |
|
|
components include:
|
| 8 |
|
|
|
| 9 |
|
|
\begin{itemize}
|
| 10 |
acosta |
1.3 |
\item Preparation of large simulated datasets (some with High Level
|
| 11 |
|
|
Trigger tags)
|
| 12 |
fisk |
1.1 |
\item Prompt reconstruction at Tier-0, including:
|
| 13 |
|
|
\begin{itemize}
|
| 14 |
|
|
\item Reconstruction at 40 Hz using CMSSW software
|
| 15 |
|
|
\item Application of calibration constants from offline DB
|
| 16 |
|
|
\item Generation of FEVT, AOD, and Alignment/Calibration skim datasets
|
| 17 |
|
|
\item Splitting of an HLT-tagged sample into O(10) streams
|
| 18 |
|
|
\end{itemize}
|
| 19 |
|
|
\item Distribution of all AOD and some FEVT to all participating Tier-1s
|
| 20 |
|
|
\item Calibration jobs on Alignment/Calibration datasets at some Tier-1s and CAF
|
| 21 |
|
|
\item Re-reconstruction performed at a Tier-1
|
| 22 |
|
|
\item Skim jobs at some Tier-1s with data propagated to Tier-2s
|
| 23 |
|
|
\item Physics jobs at Tier-2s and Tier-1s on AOD and Reco
|
| 24 |
|
|
\end{itemize}
|
| 25 |
|
|
|
| 26 |
|
|
While this is an exercise to test the data handling workflow under as
|
| 27 |
|
|
realistic conditions as possible, it is not explicitly required that
|
| 28 |
|
|
the software components are fully validated for physics performance at
|
| 29 |
|
|
the time of the challenge. However, where possible we tried to
|
| 30 |
|
|
maintain the maximum utility of the simulated, reconstructed, and
|
| 31 |
|
|
selected samples for the analysis component of the exercise. The CMS
|
| 32 |
|
|
Computing Model is described elsewhere. [Need references]
|
| 33 |
|
|
|
| 34 |
|
|
|
| 35 |
|
|
|
| 36 |
|
|
\section {Success Metric}
|
| 37 |
|
|
|
| 38 |
|
|
Success of the CSA06 challenge is pre-defined as meeting a series of
|
| 39 |
|
|
binary metrics (succeed/fail) as well as a list of quantitative
|
| 40 |
|
|
numbers, the latter of which are placed at two levels: a minimum
|
| 41 |
|
|
threshold below which we consider a definite failure, and a goal which
|
| 42 |
|
|
is considered achievable if everything runs well. No specific goals
|
| 43 |
|
|
were placed on the number or results from the calibration, alignment,
|
| 44 |
|
|
and analysis exercises.
|
| 45 |
|
|
|
| 46 |
|
|
\subsection{Binary metrics}
|
| 47 |
|
|
\begin{itemize}
|
| 48 |
|
|
\item Automatic FEVT+AOD transfer T0 to T1 via PhEDEx
|
| 49 |
|
|
\item Automatic transfer of part of FEVT+AOD T1 to T2 via PhEDEx
|
| 50 |
|
|
\item Offline DB accessible via FroNtier/Squid at participating sites
|
| 51 |
|
|
\item Insertion and use new constants in Offline DB
|
| 52 |
|
|
\item User submission of analysis/calibration/skim jobs via CRAB via DBS/DLS
|
| 53 |
|
|
\item Skim job output automatically moved to T2 via PhEDEx
|
| 54 |
|
|
\item Running re-reconstruction-like jobs at T1 that access updated information from the offline DB
|
| 55 |
|
|
\end{itemize}
|
| 56 |
|
|
|
| 57 |
|
|
\subsection {Quantitative metrics}
|
| 58 |
|
|
|
| 59 |
|
|
\begin{itemize}
|
| 60 |
|
|
\item Number of participating Tier-1 - Goal: 7 - Threshold: 5
|
| 61 |
|
|
\begin{itemize}
|
| 62 |
acosta |
1.2 |
\item Passing requires $<3$ days downtime during challenge
|
| 63 |
fisk |
1.1 |
\end{itemize}
|
| 64 |
|
|
\item Number of participating Tier-2 - Goal: 20 - Threshold 15
|
| 65 |
|
|
item Weeks of running at sustained rate - Goal: 4 - Threshold: 2
|
| 66 |
|
|
\begin{itemize}
|
| 67 |
|
|
\item This will be the period over which we measure the other metrics
|
| 68 |
|
|
\end{itemize}
|
| 69 |
|
|
\item Tier-0 Efficiency - Goal: 80\% - Threshold: 30\%
|
| 70 |
|
|
\begin{itemize}
|
| 71 |
|
|
\item Measured as unattended uptime fraction over 2 best weeks of the running period
|
| 72 |
|
|
\end{itemize}
|
| 73 |
|
|
\item Running grid jobs (T1+T2) per day (2h jobs typ.) - Goal: 50K - Threshold: 30K
|
| 74 |
|
|
\item Grid job efficiency - Goal: 90\% - Threshold: 70\%
|
| 75 |
|
|
item Data serving capability at each participating site : Goal 1MB/sec/execution slot - Threshold : 400 MB/sec (T1) or 100 MB/sec (T2)
|
| 76 |
acosta |
1.3 |
item Data transfer T0 to T1 to tape - Individual goals (threshold at
|
| 77 |
|
|
50\% of goal):
|
| 78 |
fisk |
1.1 |
\begin{itemize}
|
| 79 |
|
|
\item ASGC: 10MB/s
|
| 80 |
|
|
\item CNAF: 25 MB/s
|
| 81 |
|
|
\item FNAL: 50 MB/s
|
| 82 |
|
|
\item GridKa: 20MB/s
|
| 83 |
|
|
\item IN2P3: 25MB/s
|
| 84 |
|
|
\item PIC: 10 MB/s
|
| 85 |
|
|
\item RAL: 10MB/s
|
| 86 |
|
|
\end{itemize}
|
| 87 |
|
|
\item Data transfer T1 to T2 - Goal: 20MB/s into each T2 - Threshold: 5MB/s
|
| 88 |
|
|
\begin{itemize}
|
| 89 |
acosta |
1.3 |
\item Overall ``success'' is to have 50\% of the participants at or above goal and 90\% above the threshold
|
| 90 |
fisk |
1.1 |
\item Several T2s have better connectivity and we will have higher targets for those
|
| 91 |
|
|
\item Goal for each Tier2 is to demonstrate 50\% utilization of the WAN to the best connected T1
|
| 92 |
|
|
\begin{itemize}
|
| 93 |
|
|
\item list to be defined after SC4
|
| 94 |
|
|
\end{itemize}
|
| 95 |
|
|
\end{itemize}
|
| 96 |
|
|
\end{itemize} |
