J. Chen and B. Bershad, "The impact of operating system structure on memory system performance", Proceedings of the 16th International Symposium on Operating System Principles, December 1993, pp. 120-133.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....invocations can be reduced into the 135 cycle range on Pentiumfamily processors. Experimental analysis by Ford [8] shows that much of this performance advantage cannot be achieved in asynchronous IPC mechanisms. Given these results, performance motivated reports of the demise of microkernels [3] may have been premature. The question of security motivated demise remains open. The potential security benefit of high performance IPC is straightforward. In contrast to asynchronous or buffering IPC designs [1, 27] the latency of thread migrating IPC is low enough that applications can be ....

J. B. Chen and B. N. Bershad. The impact of operating system structure on memory system performance. In Proc. 14th Symposium on Operating Systems Principles, Dec. 1993.

....ensure the ability to interact with C and assembly components, which still form a bulk of almost all OSs out there. Secondly, if some run time infrastructure that is used to provide adaptable behaviour it very often has to ensure that it does not adversely affect the cache behaviour of the kernel [Chen, 93] Finally, many kernels that are still implemented as non paged entities [Nagle, 92] and for such systems the overall size of the kernel often becomes a hard limiting issue thus fixing the size of additional components and programming infrastructure. 1.4. Subsystem Integrity It is obvious ....

Chen, J., Bershad, B., The Impact of Operating System Structure on Memory System Performance, Proc. 14th ACM Symp. on Operating Systems Principles, Dec 93, pp. 120-133.

....performance threshold between hardware based and software based protection depends on the amount of work done in the service, but the evolution of technology will favor hardware based protection schemes. The first two assertions are not surprising. The third is different from what earlier work [Chen 93] Mogul 91] Bersh 92] implies, and is probably a result of changes in hardware architectures. The fourth is counter intuitive since it seems to say that sharing is better than copying for small data objects, but not for larger ones. The final one is most significant and, perhaps, most ....

....times have clearly not kept pace. His work did not analyze the reasons for this behavior. The impact of context switches on the cache performance is analyzed in [Mogul 91] This work demonstrated that the cost of cache refills can dominate overall cost of a context switch. Chen and Bershad [Chen 93] analyzed the effect of the system software decomposition on the memory subsystem performance. The work argues that the separation of operating system functionality into a 5 microkernel and associated user level servers significantly increases number of idle CPU cycles due to the memory stalls. ....

J. Chen, B. Bershad, "The Impact of Operating System Structure on Memory System Performance", Proc 16th Symp on Operating System Princ, ACM, pp. 120-130, 1993

....of use, we estab lished an Internet subnet for our simulated machines and entered a set of host names into the local name server. Detailed CPU simulation Workload code block Minimal binary translation ] load tnp I [lead t tplt 2, trapl 10] lw r3, 10(rl) lead add r4,r3,r2 [lo ad tmpl,simRegs[2] [ loa tmpd2, simRegs [3] t add tmp3. tmpl store tm Code annoations FWIUdata address translation PPIU data address translation Cache mmulation Cycle counting (2 instructions) Figure 4. Binary translation mechanics. Binary translation converts an instruction sequence from a block of code in ....

J.B. Chen and B. Bershad, "The Impact of Operating System Structure on Memory System Performance," Operating Symns Review, Vol. 27, No. 5, Dec. 1993, pp. 120-133.

.... executives to general purpose time sharing systems to be implemented on top of it, with the option for concurrently running other operating system personalities [HHW98] The biggest concern in using a microkernel base was the poor performance displayed by rst generation microkernels [CB93, Lie96d] The L4 microkernel [Lie95b, Lie96a, AH98] presented itself as an ideal choice not only due to its high performance implementations, but since the research group behind the PLEB project have a solid experience base with L4. This experience is due to the usage of L4 MIPS [EHL97] in ....

J. Bradley Chen and Brian N. Bershad. The impact of operating system structure on memory system performance. In Proceedings of the 14th ACM Symposium on OS Principles, pages 120-133, Asheville, NC, USA, December 1993.

....code. In 1988, Agarwal, Hennessy, and Horowitz [1] modified the microcode of the VAX 8200 to trace both user and system references and to study alternative cache organizations. Later studies were trace based. Some researchers relied on intrusive instrumentation of the OS and user level workloads [16, 48] to obtain traces; while such instrumentation can capture all memory references, it perturbs workload execution [16] Other studies employed bus monitors [26] which have the drawback of capturing only memory activity reaching the bus. To overcome this, some have used a combination of ....

....references and to study alternative cache organizations. Later studies were trace based. Some researchers relied on intrusive instrumentation of the OS and user level workloads [16, 48] to obtain traces; while such instrumentation can capture all memory references, it perturbs workload execution [16]. Other studies employed bus monitors [26] which have the drawback of capturing only memory activity reaching the bus. To overcome this, some have used a combination of instrumentation and bus monitors [78, 88, 79, 14] As an example of more recent studies, Torrellas, Gupta, and Hennessy [78] ....

CHEN, J. B., AND BERSHAD, B. N. The impact of operating system structure on memory system performance. In Proceedings of the Symposium on Operating Systems Principles (December 1993).

....the security and stability of the complete system. It extensively follows the UNIX concept of mapping all system objects into the global file name space. The performance of the Hurd is substantially impacted by Mach s slow user to user IPC and expensive user level page fault handling [CB93] Lie95] 2.2 POSIX Emulation on Windows Systems The attempts of emulating the POSIX interface on Windows are mainly targeted at reducing the e#ort necessary for porting existing programs from UNIX to Windows systems and have thus yielded a variety of products based on two fundamentally di#erent ....

J. Bradley Chen and Brian N. Bershad. The impact of operating system structure on memory system performance. In Symposium on Operating Systems Principles, pages 120--133, 1993.

.... within their distributed object environment and most code exists only at the user level; thus, Abacus is not well integrated into the kernel (e.g. one can mount an Abacusbased file system, but system calls are redirected from inside the kernel into a user level proxy, which can be inefficient [9]) One of the main strengths of the Abacus approach is that work is dynamically migrated to the client or the server depending on system and workload characteristics. A similar adaptive framework could be utilized by SRPC as well. 8.2 Extensibility SRPC is also related to a long line of work in ....

J. B. Chen and B. Bershad. The Impact of Operating System Structure on Memory System Performance. In Proceedings of the 14th ACM Symposium on Operating Systems Principles (SOSP '93), pages 120--133, Asheville, NC, December 1993.

....uses it at the same time. This problem is particularly acute for memory, which swaps data to disk when over extended. 2.2. 1 Operating Systems and Virtual Machines Page Mapping: Chen and Bershad have shown that virtual memory mapping decisions can reduce application performance by up to 50 [14]. Virtually all machines today use physical addresses in the cache tag. Unless the cache is small enough so that the page offset is not used in the cache tag, the allocation of pages in memory will affect the cache miss rate. File Layout: In [6] a simple experiment demonstrates how file system ....

J. B. Chen and B. N. Bershad. The Impact of Operating System Structure on Memory System Performance. In Proceedings of the 14th ACM Symposium on Operating Systems Principles, pages 120-- 133, December 1993.

....server are not propagated to the rest of the operating system. An erroneous server is simply terminated by the kernel. The multi server system achieves su#cient fail safety, but the performance is sacrificed because the overheads of inter process communication (IPC) and context switches are large [3, 7]. The performance of this cross domain communication has been improved 6 in recent years [22] The Chorus operating system [38, 37] allows users to download the extension modules created as the user level servers into the kernel without recompiling them. This approach achieves both su#cient ....

Chen, J. and B. N. Bershad, "The Impact of Operating System Structure on Memory System Performance," in Proceedings of the 14th Symposium on Operating System Principles, pp. 120-- 133, Dec. 1993.

....a model of seek time in shadowed disks, an early version of RAID1. Kim and Tantawi [14] analyze the positioning time delays in asynchronous disk interleaving, a variety of disk striping where sub blocks of a data block are placed independently of one another. Chen and Towsley, in several papers [4, 6], present an analytical queueing model of the response time in a disk array. Lee and Katz [15] present a model of disk striping under a simple synchronous workload. This model treats reads and writes similarly and assumes that the mean service time of a block request at a disk is known. Merchant ....

J.B. Chen and B.N. Bershad. The impact of operating system structure on memory system performance. In Proc. of 14th ACM Symp. on Operating Systems Principles (SOSP), pages 120--33, December 1993.

....compile produced data with accumulated run 6 Mach 3.0 with a Unix server was probably the first to demonstrate that a portable virtual platform can outperform even a dedicated non portable platform (Ultrix 4. 0) by using appropriate platform abstractions and by optimizing each instantiation (Chen and Bershad, 1993). time experience. The Concord project also shows how calibrated WCET functions can be obtained which provide tight WCET upper bounds for each task as a function of predetermined state variables, so that job mode bandwidths can be upgraded with each state transition. The timing analysis for each ....

Chen, J. B. and Bershad, B. N. (1993) The impact of operating system structure on memory system performance.

....servers, without lengthening the critical path of the most frequently used operations. However, performance of these first generation microkernels proved disappointing, with applications generally experiencing a significant slowdown compared to a traditional ( monolithic ) operating system [CB93] Liedtke, however, has shown [Lie93, Lie95, Lie96] that these performance problems are not inherent in the microkernel concept and can be overcome by good design and implementation. L4 is the constructive proof of this theorem, as has been clearly demonstrated by Hartig et al. HHL 97] 1.1 ....

J. Bradley Chen and Brian N. Bershad. The impact of operating system structure on memory system performance. In Proceedings of the 14th ACM Symposium on OS Principles, pages 120--133, Asheville, NC, USA, December 1993.

No context found.

J. Chen and B. Bershad, "The impact of operating system structure on memory system performance", Proceedings of the 16th International Symposium on Operating System Principles, December 1993, pp. 120-133.

No context found.

Chen, B. and Bershad, B. The impact of operating system structure on memory system performance. In Proceedings of the 14th Symposium on Operating System Principles, 1993.

No context found.

J. Bradley Chen and Brian N. Bershad. The impact of operating system structure on memory system performance. In Proc. 14th Symposium on Operating Systems Principles, December 1993.

No context found.

J. Bradley Chen and Brian N. Bershad. The impact of operating system structure on memory system performance. In 14th SOSP, pages 120--133, Asheville, NC, USA, Dec 1993. 15

No context found.

J. Bradley Chen and Brian N. Bershad. The impact of operating system structure on memory system performance. In Proceedings of the 14th ACM Symposium on OS Principles, pages 120--133, Asheville, NC, USA, December 1993.

No context found.

J. B. Chen and B. Bershad, The impact of operating system structure on memory system performance, In Proc. of 14th Symp. on Operating System Principles. ACM, 1993.

No context found.

J.B. Chen and B.N. Bershad. The impact of operating system structure on memory system performance. In Proc. of 14th ACM Symp. on Operating Systems Principles (SOSP), pages 120--33, December 1993.

No context found.

J. B. Chen and B. Bershad, The impact of operating system structure on memory system performance, In Proc. of 14th Symp. on Operating System Principles. ACM, 1993.

No context found.

Chen, J. B., and Bershad, B. N. The Impact of Operating System Structure on Memory System Performance. In Proceedings of the 14th ACM Symposium on Operating System Principles (Asheville, NC, December 1993).

No context found.

B. Bershad and J. Chen, "The Impact of Operating System Structure on Memory System Performance," The 14 Symposium on Operating System Principles, 1993.

No context found.

J. Bradley Chen and Brian N. Bershad. The Impact of Operating System Structure on Memory System Performance. In Proceedings of the 14th ACM Symposium on Operating Systems Principles, pages 120--133, December 1993.

No context found.

J. B. Chen, and B. N. Bershad, "The impact of operating system structure on memory system performance," Proc. of the ACM Symposium on Operating System Principles, pp. 120-133, Dec. 1993.

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