David Kotz and Preston Crow. The expected lifetime of single-address-space operating systems. Computing Systems, 9(3):155--178, Summer 1996.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....a cluster operating system, targeted at parallel applications. Its naming scheme decouples naming from storage, a characteristic we regard as important for scalable single system images [FCH 93] Kotz and Crow examined how SASOS designs depend on hardware with su#cient bits of address space [KC96] Ansley and Ashton discuss the challenges of administering a SASOS system; specifically the impact of a distributed memory on resiliency in the event of component failure [AA96] All of the systems described above unify the resources of a cluster of machines to provide a single system image. In ....

David Kotz and Preston Crow. The expected lifetime of single-address-space operating systems. Computing Systems, 9(3):155--178, Summer 1996.

....passwords than the old one. Hence dangling pointers and capabilities do not present a security problem, although they have other problems similar to those of dangling symbolic links in UNIX. Address space reuse is important as otherwise even a 64 bit address space could conceivably be exhausted [23]. With reuse, address space consumption is essentially limited by the amount of backing store available, which ensures that a 64 bit address space will suffice until it becomes feasible to connect billions of gigabytes of disk to a single system. Object table All information about objects, ....

D. Kotz and P. Crow. The expected lifetime of single-address-space operating systems. Computing Systems, 9:155--178, 1996.

....each entry translates. This creates challenges in the efficient management and use of large address spaces [16, 11, 15] As TLB coverage struggles, or becomes unable to map entire process working sets, MMU address translation performance becomes a very real limitation on real system performance [14]. Cache performance studies are normally conducted experimentally. Address translation evaluation techniques generally utilise a software simulator [2, 5] because of the cost, complexity and difficulty of exact physical measurements from hardware prototypes. We have developed a simulation package ....

D Kotz and P Crow. The expected lifetime of Single-Address-Space operating systems. ACM SIGMETRICS, 22(1):161--170, 1994.

....persistent objects. As a matter of fact, applications are becoming more and more complex, and their object graphs extremely intricate. Thus, manual storage management is increasingly difficult and error prone. Even in 64 bit address spaces, memory reorganization and address recycling are necessary [12]. Otherwise, garbage ends up filling the secondary storage, and the address space becomes fragmented. Our GC design is integrated in a platform, called BMX (Bunch Manager Executive) 6] supporting distributed persistent objects, not only via a weakly This article appears in the Proceedings of ....

David Kotz and Preston Crow. The expected lifetime of single-address-space operating systems. In Proceedings of SIGMETRICS'94, Nashville, Tennessee, (USA), May 1994. ACM Press.

....each entry translates. This creates challenges in the efficient management and use of large address spaces [21, 12, 19] As TLB coverage struggles, or becomes unable to map entire process working sets, MMU address translation performance becomes a very real limitation on real system performance [18]. Recent TLB and cache performance studies indicate that CPU architects perceive a real problem in this area. A number of researchers have proposed alternative ideas to the traditional forward mapped page table combined with a fixed page size design. Several ideas have been introduced to ....

D Kotz and P Crow. The expected lifetime of Single-Address-Space operating systems. ACM SIGMETRICS, pages 161--170, 1994.

....of the pages that each TLB entry translates. Thus the amount of memory instantaneously visible at an instant, termed TLB coverage, now struggles, or has become unable to map entire process working sets. Therefore, address translation performance becomes a very real limitation on system performance [Kotz and Crow, 1994]. Talluri s investigation of techniques designed to increase TLB reach [Talluri, 1995] concluded that partial subblocking [Talluri and Hill, 1994] offered the best compromise between superpages [Khalidi et al. 1993; Talluri et al. 1992] and fully subblocked [Talluri and Hill, 1994] TLB ....

Kotz, D and Crow, P (1994). The expected lifetime of Single-Address-Space operating systems. ACM SIGMETRICS, 22(1):161--170.

....each entry translates. This creates challenges in the efficient management and use of large address spaces [19, 12, 17] As TLB coverage struggles, or becomes unable to map entire process working sets, MMU address translation performance becomes a very real limitation on real system performance [16]. Large virtual and real memory address have increased the number of pages in both virtual and physical memory. The number of physical pages on a large computer system is approaching 1,000,000. The management of such large numbers of pages becomes extremely unwieldy and inefficient. Our contention ....

D Kotz and P Crow. The expected lifetime of Single-Address-Space operating systems. ACM SIGMETRICS, 22(1):161--170, 1994.

....each entry translates. This creates challenges in the efficient management and use of large address spaces [19, 10, 17] As TLB coverage struggles, or becomes unable to map entire process working sets, MMU address translation performance becomes a very real limitation on real system performance [16]. Recent TLB and cache performance studies indicate that CPU architects perceive a real problem in this area. A number of researchers have proposed alternative ideas to the traditional forward mapped page table combined with a fixed page size design. Several ideas have been introduced to ....

D Kotz and P Crow. The expected lifetime of Single-Address-Space operating systems. ACM SIGMETRICS, pages 161--170, 1994.

.... Mungi, and Brevix [CLFL94, MWSK94, HEV 98, FCH 93] The SASOS architecture is a useful form of transparency, but to scale across arbitrary administrative domains a single address space would have to span the global Internet; that would introduce allocation and administrative problems [KC96] Other systems aim to provide a world wide distributed system. The worldwide approaches involve aggregating all resources into one large system, then partitioning that system among its users. Such a system would provide a single system image across all administrative domains for all users. ....

David Kotz and Preston Crow. The expected lifetime of single-address-space operating systems. Computing Systems, 9(3):155--178, Summer 1996.

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