Abstract

As useful as performance counters are, the meaning of reported aggregate event counts is sometimes questionable. Questions arise due to unanticipated processor behavior, overhead associated with the interface, the granularity of the monitored code, hardware errors, and lack of standards with respect to event definitions. To explore these issues, we are conducting a sequence of studies using carefully crafted microbenchmarks that permit the accurate prediction of event counts and investigation of the differences between hardware-reported and predicted event counts. This paper presents the methodology employed, some of the microbenchmarks developed, and some of the information uncovered to date. The information provided by this work allows application developers to better understand the data provided by hardware performance counters and better utilize it to tune application performance. A goal of this research is to develop a cross-platform microbenchmark suite that can be used by application developers for these purposes. Some of the microbenchmarks in this suite are discussed in the paper.

As useful as performance counters are, the meaning of reported aggregate event counts is sometimes questionable. Questions arise due to unanticipated processor behavior, overhead associated with the interface, the granularity of the monitored code, hardware errors, and lack of standards with respect to event definitions. To explore these issues, we are conducting a sequence of studies using carefully crafted microbenchmarks that permit the accurate prediction of event counts and investigation of the differences between hardware-reported and predicted event counts. This paper presents the methodology employed, some of the microbenchmarks developed, and some of the information uncovered to date. The information provided by this work allows application developers to better understand the data provided by hardware performance counters and better utilize it to tune application performance. A goal of this research is to develop a cross-platform microbenchmark suite that can be used by application developers for these purposes. Some of the microbenchmarks in this suite are discussed in the paper.

This project focuses on building a reservoir sub-sea network model for a condensate field in the gulf of Guinea, the Duke Field. It integrates the five developed Duke reservoirs, development wells and subsea network using the Petroleum Experts' Integrated Production Model suite of software, (IPM) which is widely used in the E&P industry especially for integrated forecasting, surveillance and production system optimization that require integration of surface and subsurface models.Following the acquisition and quality control of data from other teams working on the Duke Field, a network model which integrates the five Duke reservoirs, their associated wells and subsea network up to the production separator was built. The model was initialized and used to predict full field performance under different scenarios.Finally, a water injection allocation sensitivity study was performed and the results were analyzed both technically and economically. From the technical point of view, the option to reallocate 10 kbwpd from reservoir U to reservoir P-upper North and another 10 kbwpd from Reservoir ST to reservoir Q-Lower brought about the optimum recovery. This was also supported by a simple economic analysis. It was then recommended that additional water injectors be drilled in P-Upper North and Q-Lower to unlock an additional 8.4 MMSTB of reserves resulting from higher sweep efficiencies and better pressure maintenance.