Integrated Tracker Task Force Review

Introduction

On September 23^rd/24^th 2003, the ITTF work was presented to a review panel composed of representatives from several detector-sub-systems experts, tracking experts from within and outside the collaboration, Software and Physics experts. The review committee panel endorsed the STI tracker as the future tracking Software for STAR. However, a follow up review was requested to finalize and clearly demonstrate several aspects of the tracker which, in the opinion of the panel, was not clearly or un-ambiguously demonstrated. The following main points needed to be better addressed :

1. The design flexibility was in-sufficiently presented. Especially, the integration of the geometry for a new detector sub-system was not adequately demonstrated. A comparison of material between GEANT and the STI implementation was requested as a validation of the STI geometry approach. Comparative radiation plots were suggested.

2. Track extrapolation to other volumes (BEMC and EEMC for example) were not shown. Further integration of shower max detector (SMD) hits was requested or demonstrated to be feasible.

3. Comparisons between TPT/EST and STI/SVT were requested.

4. Treatment of energy loss and multiple coulomb scattering should be re-assessed ; some of those effects were not enabled at the preceding review.

5. At the preceding review, mainly due to a lack of time, the results were inconsistent. A clear demonstration of equal or better performance than the present STAR tracking code is the basic requirement made by the review committee. A particular attention to the low Pt region was stressed. The presented HBT results were inconclusive and needed to be re-assessed.

6. The hit error parametrization needed clarifications and its effect on tracking performances explained.

7. A reliable method for improving the tracker performances was requested and was a requirement for a later better tuning of the tracker's efficiencies.

A new review date was proposed as for December 2002 with a deployment in early 2003 and a double pass production using both trackers for the d+Au run. The committee has further pointed at the need of the deployment of the new tracker before the year 4 data taking and the impact on the Physics capabilities of STAR if such deployment would not occur. However, due to Physics program constraints and manpower availability, a deployment plan and work toward this second review was not possible at the time. At this stage, we believe the ITTF core developers have benefited from the extra time to strengthen the tracker performance and its design.

Current Charges

The initial committee charges are provided in appendix along with the review panel conclusions. We would like to present to the current committee the following charges which are complementary to the preceding review. The review committee should be presented enough documentation and demonstration which would lead to conclude and bring their final comments and recommendations on the following issues :

1. The integration feasibility of other detector sub-systems, track extension to volume and integration of hit-driven detectors should be clarified and/or clearly demonstrated.

2. The code design should be sufficiently explained for allowing the committee to comments on its flexibility. Implementation of
   - other track models or seed finders
   - and/or the possibility of running concurrent models
   - and the flexibility of making multiple call to the tracking routines (multiple passes)
   should be clarified at minimum through provided documentation. Documentation on the design should be complete, clear and self-explanatory

3. For the geometry approach and design choice, suggested comparisons between GEANT and STI (or equivalent validation of the geometry model) are expected.

4. The energy loss and coulomb scattering treatment should be presented and address the committee's concerns and request for clarification. The influence of those effects and the parametrization of hit errors on the quality of the tracker should be understood.

5. STI tracker's efficiencies and global tracking performance must be presented using the same set of data and conditions : primary and secondaries, from Monte Carlo tracks, embedding and real data under low and high occupancy conditions. All results should be compared to the currently used STAR tracker as a base line performance. The nature of the tracking inefficiencies, if subsist, should be identified and explained. However, as the current tuning of the STI tracker may not be final, we would like the review panel to allow for a “equivalent efficiencies” requirement.

6. Performances for TPT/EST compared to STI/SVT should be presented. This would clearly demonstrate the power of the new tracker.

7. The integration in the STAR framework and a deployment plan was necessary for the committee to approve its use by the STAR collaboration. As suggested by the committee in the past, we would like to address this issue separately as outlined above in a second part of the charges.

Post review considerations

Shall the STI tracker be approved at this stage by the committee:

Recognizing that a successful transition to a new framework goes beyond Software & Computing, the committee is further asked for advise and/or make clear recommendations on how to achieve the integration of STAR to the STI tracker. In particular, and considering the rapid approach toward the Year 4 run :

1. The committee is aware of both the currently used STAR tracker slow decay and its lack of quality and flexibility. Does the committee see the STI tracker as being a key component to our Year 4 Physics Analysis success ? If so, could the committee comment on what, in their mind, is the most convincing advantage of the STI tracker and what impact would a delay in using it have on our Physics program and/or capabilities ?

2. Should the current TPT/EGR tracker and related developments be abandoned and manpower re-routed to the STI integration ? If the consensus is for dropping TPT support, when would be a reasonable time for doing so ? Shall the Year4 data be processed with both tracker for comparison purposes or a MDC5-approach (both involving real data and MC data) be preferable and sufficient ?

3. The preceding committee suggested the allocation of one person per detector sub-system and two per Physics working group for testing, calibration and integration purposes. Would the committee still see this allocation as being needed ? Considering the time available between the review and the Year 4 run, could the committee advise on a time window or a realistic time scale for such manpower concentration and involvement ?

4. Could the committee re-assess / suggest / recommend / comment on how to ensure the long term support of the STI tracker ?

Shall the STI tracker need further development according to the committee:

Due to the urgency for a new tracker, the committee is asked to consider the above questions as a rough guideline for an executive session discussion.

1. What is the acceptable time scale for presenting further information and/or documentation to the committee ?

2. Shall extra manpower be taken / invested to complete the project ? What are our options to achieve our goals ?

3. Any comments / suggestions or recommendations on the road map in regard of the Year 4 approach ?

Finally, the review committee is asked to return their written report within two weeks following the review to the STAR Software and Computing leader.

The review committee

The review committee will be composed of the following members :