Perl, S. E. and Sites, R. L. (1996). Studies of Windows NT Performance Using Dynamic Execution Traces. In Proc. of the Second Symposium on Operating Systems Design and Implementation, pages 169--184.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....is to increase the amount of cache available on chip. For some applications, the on chip L1 caches will suffice [Lee 98] For other applications such as commercial or database programs, the working data set is too large to fit in the L2 cache and will still need to access main memory [Maynard 94, Perl 96, Barroso 98] Note that the L1 cache size is constrained by the desire to have a small access latency and cannot be made arbitrarily large. Similar in spirit to increasing the amount of cache available on chip are designs that integrate microarchitectures and main memory. The idea is to put a ....

....cost effective. Another current trend is that caches are increasing their geometry in size and set associativity. While some smaller applications can be accommodated with relatively small L1 caches [Romer 96, Lee 98] many applications still exist that do not fit in larger L2 caches [Maynard 94, Perl 96] making research on how to efficiently use caches relevant. As caches become larger and more set associative, strategies that emphasize the reuse of lines already present in the cache might be more important than those that specifically target less associative caches. The increase in cache set ....

S. E. Perl and R. L. Sites. Studies of Windows NT Performance Using Dynamic Execution Traces. In 2nd USENIX Symposium on Operating Systems Design and Implementation, pages 169--183, October 1996.

.... database systems running on line transaction processing (OLTP) benchmarks incur high rates of cache misses and instruction stalls, which re duce processor performance to as low as a tenth of its peak potential [4, 9, 20] Part of this problem may be attributable to database systems code size [28], but their execution model is also responsible. These systems are structured so that a process or thread runs for a short period before invoking a blocking operation and relinquishing control, so processors execute a succession of diverse, non looping code segments that exhibit little locality. ....

Sharon Perl and Richard L. Sites, "Studies of Windows NT Performance using Dynamic Execution Traces," in Proceedings of the Second USENIX Symposium on Operating Systems Design and Implementation (OSDI). Seattle, WA, October 1997, pp. 169-183.

....processing, business decision support, and e business applications are driving the development of powerful server systems. From an execution point of view, commercial workloads are different from technical workloads and present more vigorous demands on the memory and storage sub systems [1, 2, 3]. In fact, studies [4] that analyzed transaction processing workloads indicate that systems spend a significant fraction of the execution time waiting for I O devices to access the data. Once the data is brought to main memory, the processor uses a substantial amount of the remaining execution ....

S. E. Perl and R. L. Sites, "Studies of Windows NT performance using dynamic execution traces," in Proceedings of the 2nd USENIX Symposium on Operating Systems Design and Implementation, (Seattle, WA, USA), pp. 169-184, Oct. 28-31 1996. 17

....tend to be used by most users. 1.2 Related Work Although the majority of today s personal computers run mostly interactive applications on Windows systems, there has been little research on how people use these programs. Several research projects investigated Windows operating system performance [CE96, EW96, PS96]. A recent paper [LC98] presented measurements and simulation results of instruction set and architectural characteristics during program execution on x86 Windows NT. These projects focused on the characteristics and comparison of the general system performance, while this paper focuses on the ....

S. Perl and R. Sites, "Studies of Windows NT Performance Using Dynamic Execution Traces." In Proceedings of the Second Symposium on Operating Systems Design and Implementation, USENIX Association, pages 169-183, October 1996.

....first to study the performance of memory consistency models in the context of database workloads. There are a number of studies based on the performance of out of order processors for non database workloads(e.g. 42, 85, 89] Most previous studies of databases are based on in order processors [9, 27, 28, 34, 75, 94, 118, 119], and therefore do not address the benefits of more aggressive processor architectures. A number of the studies are limited to 102 uniprocessor systems [27, 34, 72, 75] As discussed in Section 4.4, data communication misses play a more dominant role in multiprocessor executions and somewhat ....

....executions and somewhat reduce the relative effect of instruction stall times. Another important distinction among the database studies is whether they are based on monitoring existing systems [3, 13, 18, 26, 27, 28, 57, 62, 118] typically through performance counters) or based on simulations [7, 8, 34, 72, 75, 82, 94, 106, 119, 120]; one study uses a combination of both techniques [9] Monitoring studies have the advantage of using larger scale versions of the workloads and allowing for a larger number of experiments to be run. However, monitoring studies are often limited by the configuration parameters of the system and ....

Sharon E. Perl and Richard L. Sites. Studies of Windows NT Performance using Dynamic Execution Traces. In Proceedings of the Second Symposium on Operating System Design and Implementation, pages 169--184, October 1996.

....assembly code. Thus, the ability to compile programs to a small representation is important to reduce both software development costs and manufacturing costs. High performance systems are also impacted by program size due to the delays incurred by instruction cache misses. A study at Digital [Perl96] measured the performance of an SQL server on a DEC 21064 Alpha. Due to instruction cache misses, the application could have used twice as much instruction bandwidth as the processor was able to provide. This problem is exacerbated by the growing gap between the cycle time of microprocessors and ....

S. Perl and R. Sites, "Studies of Windows NT Performance Using Dynamic Execution Traces", Proceedings of the USENIX 2nd Symposium on Operating Systems Design and Implementation, pp. 169-183, October 1996.

....that memory bandwidth limitations are responsible for large performance degradations in many important applications. For example, Burger et al. [2] report between 11 and 31 of the total memory stalls observed in several SPEC benchmarks are due to insu#cient memory bandwidth. Perl and Sites [12] find the memory bandwidth required by a commercial SQL Server exceeds the pin bandwidth provided by the Alpha 21164 by a factor of 2. Ding and Kennedy [5] observe several scientific kernels require between 3.4 and 10.5 times the L2 memory bus bandwidth provided by the SGI Origin 2000. While ....

Sharon E. Perl and Richard L. Sites. Studies of Windows NT Performance Using Dynamic Execution Traces. In Proceedings of the Second USENIX Symposium on Operating Systems Design and Implementation, October 1996.

....servers. While applications such as decision support (DSS) and Web index search have been shown to be relatively insensitive to memory system performance [2] a number of recent studies have underscored the radically different behavior of online transaction processing (OLTP) workloads [2, 6, 7, 15, 17, 19, 25]. In general, OLTP workloads lead to inefficient executions with a large memory stall component and present a more challenging set of requirements for processor and memory system design. This behavior arises from large instruction and data footprints and high communication miss rates that are ....

S. E. Perl and R. L. Sites. Studies of windows NT performance using dynamic execution traces. In Proceedings of the Second Symposium on Operating System Design and Implementation, pages 169-- 184, Oct. 1996.

.... the resource usage patterns of desktop and commercial applications running on Windows by instrumenting applications [16, 17] Perl and Sites studied the performance of databases, compilers, and scientific codes on DEC Alpha machines running Windows NT using instruction and memory reference traces [25]. WatchTower may be helpful for studies such as these, providing a simple way to produce traces. Conversely, studies such of these can inform the development of Argus s data reduction algorithms. 6 Conclusions We have described WatchTower, a system that simplifies the collection of Windows ....

PERL, S. E., AND SITES, R. L. Studies of windows nt performance using dynamic execution traces. In Proceedings of the Second Symposium on Operating Systems Design and Implementation (OSDI '96) (October 1996), pp. 169-- 184.

....these checkers, OpenBSD is always worse than Linux, ranging from about 20 worse to almost a factor of six. spection of error logs, 2) analysis of system behavior under fault injection, and (3) testing. We consider each below. While there have been many performance studies of operating systems [5, 21, 20], there have been relatively few that look at code quality from within the OS community. Most defect studies come from the software engineering or fault tolerant elds, and almost all are based on data gathered from post mortem inspection of error logs or defect reports, typically for high ....

S.E. Perl and R.L. Sites. Studies of Windows NT Performance Using Dynamic Execution Traces. In Operating Systems Design and Implementation, pages 169-183, 1996.

....these checkers, OpenBSD is always worse than Linux, ranging from about 20 worse to almost a factor of six. spection of error logs, 2) analysis of system behavior under fault injection, and (3) testing. We consider each below. While there have been many performance studies of operating systems [5, 21, 20], there have been relatively few that look at code quality from within the OS community. Most defect studies come from the software engineering or fault tolerant fields, and almost all are based on data gathered from post mortem inspection of error logs or defect reports, typically for high ....

S.E. Perl and R.L. Sites. Studies of Windows NT Performance Using Dynamic Execution Traces. In Operating Systems Design and Implementation, pages 169--183, 1996.

....these checkers, OpenBSD is always worse than Linux, ranging from about 20 worse to almost a factor of six. spection of error logs, 2) analysis of system behavior under fault injection, and (3) testing. We consider each below. While there have been many performance studies of operating systems [5, 21, 20], there have been relatively few that look at code quality from within the OS community. Most defect studies come from the software engineering or fault tolerant fields, and almost all are based on data gathered from post mortem inspection of error logs or defect reports, typically for high ....

S.E. Perl and R.L. Sites. Studies of Windows NT Performance Using Dynamic Execution Traces. In Operating Systems Design and Implementation, pages 169--183, 1996.

....of this server include symmetric multiprocessor systems (SMP) as well as cluster servers. When we look at the execution behavior of commercial workloads, we observe commercial workloads are di erent than technical workloads and present more vigorous demands to the memory and storage sub systems [3, 4, 5]. In fact, studies that analyzed transaction processing workloads indicate that systems spend around 90 of the time waiting for the I O devices to access the data [6] Once the data are brought to memory, the processor uses between 25 to 45 of the execution time handling memory accesses [7] ....

S. E. Perl and R. L. Sites, \Studies of Windows NT Performance Using Dynamic Execution Trees," in Proceedings of the 2nd USENIX Symposium on Operating Systems Design and Implementation, (Seattle, WA, USA), pp. 169-184, Oct. 28-31 1996.

....transaction processing, decision support, microbenchmark, and performance evaluation. 1. INTRODUCTION In the last five to ten years, several studies have explored the architectural characteristics of online transaction processing (OLTP) database workloads [3] 7] 8] 9] 16] 17] 18] 19] [22] [23] 24] 26] 27] 1 This work was performed as part of the author s dissertation research. The author s present address is: Storage Systems Program, Hewlett Packard Laboratories, 1501 Page Mill Road, M S 1U 13, Palo Alto, CA 94304 1126. Her current email address is kkeeton hpl.hp.com. 50 ....

....most promising approach for producing a representative random microbenchmark lies in posing multiple read only queries. 5. RELATED WORK Many of the studies that use database workloads to evaluate computer architecture innovations have employed the complex OLTP [3] 7] 8] 9] 16] 17] 18] 19] [22] [23] 24] 26] 27] 33] and DSS [3] 5] 15] 17] 18] 23] 31] workloads defined by the TPC. These studies vary in their usage of full scale data sets versus in memory data sets. One study provides rules of thumb for using an in memory version of the TPC B OLTP benchmark to approximate the ....

S. E. Perl and R. L. Sites. "Studies of windows NT performance using dynamic execution traces," In Proc. of the Second USENIX Symposium on Operating Systems Design and Implementation (OSDI), pages 169-184, 1996.

.... used a proprietary software instrumentation tool to examine the performance of various technical and commercial workloads, including TPC A and TPC C, on the IBM RS 6000, which employs a Pow 59 Characteristics Maynard94 [64] Cvetanovic94 [27] Cvetanovic96a [28] Rosenblum94 [87] Perl96 [80] System evaluated RS 6000 DEC 7000 AXP AlphaServer 8200 SGI based UP SMP Alpha PCs, 4 proc. AlphaServer 2100 Processor PowerPC (in order) Alpha 21064 (in order) Alpha 21164 (in order) MIPS like (UP: out of order, MP: in ord. Alpha 21064, 21164 (in order) I O system UP: 2 ....

....and extra instructions misses for different data structures. 3.8.1.4. Perl and Sites: Alpha PCs Perl and Sites used the PatchWrx tracing facility to gather traces of several technical workloads and a TPC B based OLTP workload run by Microsoft SQL Server on a Windows NT based Alpha 21064 PC [80]. They examined the processor pin bandwidth required for these workloads for two uniprocessor designs, based on the 21064 the 21164. This examination is based on an estimation of the bandwidth required to 63 achieve a CPI of 1, and comparison of the estimated bandwidth to the actual bandwidth ....

S. E. Perl and R. L. Sites. "Studies of windows NT performance using dynamic execution traces," Proc. of the Second USENIX Symposium on Operating Systems Design and Implementation (OSDI), pages 169-184, 1996.

No context found.

Perl, S. E. and Sites, R. L. (1996). Studies of Windows NT Performance Using Dynamic Execution Traces. In Proc. of the Second Symposium on Operating Systems Design and Implementation, pages 169--184.

No context found.

PERL, S. E., AND SITES, R. L. Studies of windows nt performance using dynamic execution traces. In Proceedings of the Second Symposium on Operating Systems Design and Implementation (OSDI '96) (October 1996), pp. 169-- 184.

No context found.

S. E. Perl and R. L. Sites. "Studies of windows NT performance using dynamic execution traces," In Proc. of the Sec- ond USENIX Symposium on Operating Systems Design and Implementation, pages 169-184, 1996.

No context found.

S. Perl and R. Sites, "Studies of Windows NT Performance Using Dynamic Execution Traces", in ACM SIGOPS Operating System Reviews, V.30, No.10, Oct. 1996, pp 169183.

No context found.

Sharon E. Perl, Richard L. Sites, "Studies of Windows NT Performance using Dynamic Execution Traces", 1996.

No context found.

S. E. Perl and R. L. Sites. Studies of Windows NT performance using dynamic execution traces. In Proceedings of the Second USENIX Symposium on Operating Systems Design and Implementation, pages 169-183, Seattle, WA USA, Oct. 1996.

No context found.

S. E. Perl and R. L. Sites. Studies of Windows NT performance using dynamic execution traces. In Second USENIX Symposium on Operating System Design and Implementation, pages 169--183, October 1996.

No context found.

S.E. Perl and R.L. Sites, "Studies of Windows NT Performance Using Dynamic Execution Traces," Digital Systems Research Center Research Report, RR-146, April 1997.

No context found.

S. E. Perl and R. Sites. Studies of Windows NT performance using dynamic execution traces. In Proceedings of the Second Symposium on Operating System Design and Implementation, pages 169#184, October 1996.

No context found.

S.E. Perl and R.L. Sites, "Studies of Windows NT Performance Using Dynamic Execution Traces," Proc. Second Symp. Operating Systems Design and Implementation, 1996, pp. 169-183.

First 50 documents  Next 50

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC