Disk arrays provide the raw storage throughput needed to balance rapidly increasing processor performance. Unfortunately, many important, I/O-intensive applications have serial I/O workloads that do not benefit from array parallelism. The performance of a single disk remains a bottleneck on overall performance for these applications. In this dissertation, I present aggressive, proactive mechanisms that tailor file-system resource management to the needs of I/O-intensive applications. In particular, I will show how to use applicationdisclosed access patterns (hints) to expose and exploit I/O parallelism, and to dynamically allocate file buffers among three competing demands: prefetching hinted blocks, caching hinted blocks for reuse, and caching recently used data for unhinted accesses. My approach estimates the impact of alternative buffer allocations on application elapsed time and applies run-time cost-benefit analysis to allocate buffers where they will have the greatest impact. I implemented TIP, an informed prefetching and caching manager, in the Digital UNIX operating system and measured its performance on a 175 MHz Digital Alpha workstation equipped with up to 10 disks running a range of applications. Informed prefetching on a ten-disk array reduces the wall-clock elapsed time of computational physics, text search, scientific visualization, relational database queries, speech recognition, and object linking by 10-84% with an average of 63%. On a single disk, where storage parallelism is unavailable and avoiding disk accesses is most beneficial, informed caching reduces the elapsed time of these same applications by up to 36% with an average of 13% compared to informed prefetching alone. Moreover, applied to multiprogrammed, I/O-intensive workloads, TIP increases overall throughput. I am particularly indebted to Jim Zelenka for his tireless efforts debugging and improving the raw TIP code I threw over the wall to him during the crush to produce the SOSP paper. He has an uncanny ability to track down kernel bugs which I gave him many opportunities to exercise. I am fortunate to have both benefited and learned from this skill. Writing a dissertation requires more than just research, and I am grateful to the many folks who helped me keep my senses of perspective and humor over the years. Peter Stout, besides answering my endless computer questions, provided a fine example of how it should be done. He also taught me how to drink coffee in large quantities. In the later years, David Steere and Brian Noble were always ready with a humorous diversion even when the slogging seemed slow and endless. Thanks too to Patty Mackiewicz who helped with many practical problems and always looked out for me. But, more importantly, she found a way to make me smile even in the face of adversity.
The survey data indicate that there is a considerable variation in the management of patients with severe head injury in the United States. The establishment of guidelines for the management of head injury based on available scientific data and moderated by practical and financial considerations may lead to improvement in the standard of care.
Three cases of thoracic disc herniation presenting with signs of spinal cord compression are reported. The patients were operated on by an approach through a midline incision in which a pedicle is removed. Two patients were cured and one has improved.
Disk arrays provide the raw storage throughput needed to balance rapidly increasing processor performance. Unfortunately, many important, I/O-intensive applications have serial I/O workloads that do not benefit from array parallelism. The performance of a single disk remains a bottleneck on overall performance for these applications. In this dissertation, I present aggressive, proactive mechanisms that tailor file-system resource management to the needs of I/O-intensive applications. In particular, I will show how to use applicationdisclosed access patterns (hints) to expose and exploit I/O parallelism, and to dynamically allocate file buffers among three competing demands: prefetching hinted blocks, caching hinted blocks for reuse, and caching recently used data for unhinted accesses. My approach estimates the impact of alternative buffer allocations on application elapsed time and applies run-time cost-benefit analysis to allocate buffers where they will have the greatest impact. I implemented TIP, an informed prefetching and caching manager, in the Digital UNIX operating system and measured its performance on a 175 MHz Digital Alpha workstation equipped with up to 10 disks running a range of applications. Informed prefetching on a ten-disk array reduces the wall-clock elapsed time of computational physics, text search, scientific visualization, relational database queries, speech recognition, and object linking by 10-84% with an average of 63%. On a single disk, where storage parallelism is unavailable and avoiding disk accesses is most beneficial, informed caching reduces the elapsed time of these same applications by up to 36% with an average of 13% compared to informed prefetching alone. Moreover, applied to multiprogrammed, I/O-intensive workloads, TIP increases overall throughput. I am particularly indebted to Jim Zelenka for his tireless efforts debugging and improving the raw TIP code I threw over the wall to him during the crush to produce the SOSP paper. He has an uncanny ability to track down kernel bugs which I gave him many opportunities to exercise. I am fortunate to have both benefited and learned from this skill. Writing a dissertation requires more than just research, and I am grateful to the many folks who helped me keep my senses of perspective and humor over the years. Peter Stout, besides answering my endless computer questions, provided a fine example of how it should be done. He also taught me how to drink coffee in large quantities. In the later years, David Steere and Brian Noble were always ready with a humorous diversion even when the slogging seemed slow and endless. Thanks too to Patty Mackiewicz who helped with many practical problems and always looked out for me. But, more importantly, she found a way to make me smile even in the face of adversity.
✓ The records of 51 patients with craniopharyngiomas were reviewed to compare the clinical features of the tumor in children and adults, to weigh the results of radical removal of the neoplasm, and to evaluate the benefit of postoperative radiation therapy. Signs and symptoms in the two age groups differed modestly. Endocrine abnormalities and visual field deficits were recognized more often in adults, whereas skull films were more likely to be abnormal in children. Six patients underwent radical removal of tumor. All have returned to normal life and remain free of recurrence. When tumor removal was incomplete, recurrence was common, appearing earlier in the younger patients. The administration of postoperative radiotherapy following partial removal of tumor appeared to delay recurrence in both age groups.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.