Wang, R. Y., Anderson, T. E., and Patterson, D. A. (1998b). Virtual Log Based File Systems for a Programmable Disk. Technical Report UCB/CSD 98/1031, University of California at Berkeley.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....performance is limited by mechanical delays. In order to reduce disk I O overheads, several techniques and optimizations have been proposed. This literature includes work on optimizing the scheduling of disk requests [33, 3, 17, 15] disk arrays [10, 4, 34] and optimizing disk writes using logs [23, 13, 32]. Despite these techniques and optimizations, low disk throughput is still a serious problem for data intensive servers, such as Web proxies, email and news servers, multimedia servers, and database servers. An important reason for this is that current disk controller caches are not designed ....

....There have been several works on improving disk I O performance. Most of them discuss techniques that are external to the disk drive. Typical examples are techniques for improving the prefetching and caching of disk blocks [21, 6, 27] scheduling requests [3, 15] and for optimizing disk writes [13, 23, 32]. Our techniques deal with read ahead (a form of prefetching) and caching of disk data, but their focus is on the disk controller cache. There are several important differences between the controller cache and other caches, such as the buffer cache, the Disk Caching Disk [13] or even a ....

R. Y. Wang, T. E. Anderson, and D. A. Patterson. Virtual Log Based File Systems for a Programmable Disk. In Proceedings of the 3th Symposium on Operating Systems Design and Implementation, pages 29--44, New Orleans, LA, February 1999.

....System [58] 59] writes data to disk sequentially in a log like structure. The log contains indexing information so that file data can be read back from the log efficiently. Another way to minimize disk head movement is to write to a disk location that is close to the current disk head location [72]. There are also hybrid approaches that combine NVRAM with disk head minimizing techniques during writes. In DCD [30] disk writes are first staged to NVRAM and later written sequentially to a cache disk. The data in the cache disk is subsequently updated in place in the data disk. The two level ....

R.Y. Wang, T.E. Anderson, and D.A. Patterson, "Virtual Log Based File Systems for a Programmable Disk," Proc. 1995.

.... To know that we know what we know, and that we do not know what we do not know, that is true knowledge. Confucius As microprocessors and memory chips become smaller, faster, and cheaper, embedding processing and memory in peripheral devices has become an increasingly attractive proposition [1, 19, 32, 40]. Placing processing power and memory capacity within a smart disk system allows functionality to be migrated from the file system into the disk (or RAID) thus providing a number of potential advantages over a traditional system. For example, when computation takes place near data, one can ....

....directories, and indirect blocks) are not exposed. Thus, research efforts have been limited to applying disk system intelligence in a manner that is oblivious to the nature and meaning of file system traffic, e.g. improving write performance by writing blocks to the closest free space on disk [15, 40]. To fulfill their potential and retain their utility, smart disk systems must become smarter while the interface to storage remains the same. Such a system must acquire knowledge of how the file system is using it, and exploit that understanding in order to enhance functionality or increase ....

[Article contains additional citation context not shown here]

R. Wang, T. E. Anderson, and D. A. Patterson. Virtual Log-Based File Systems for a Programmable Disk. In Proceedings of the Third Symposium on Operating Systems Design and Implementation (OSDI '99), New Orleans, LA, February 1999.

....almost all such techniques have been developed under the assumption that the file system will be run upon a single, traditional disk. More recently, storage systems have also received much attention. For example, smart disks can improve read or write performance with block remapping techniques [11, 13, 49]. For I O intensive workloads, multiple disk storage systems have been well studied in the research community [26, 51] and have achieved success in the storage industry. These high end storage systems provide the illusion of a single, fast disk to unsuspecting file systems above, but internally ....

R. Wang, T. E. Anderson, and D. A. Patterson. Virtual LogBased File Systems for a Programmable Disk. In Proceedings of the Third Symposium on Operating Systems Design and Implementation (OSDI '99), New Orleans, LA, February 1999.

.... To know that we know what we know, and that we do not know what we do not know, that is true knowledge. Confucius As microprocessors and memory chips become smaller, faster, and cheaper, embedding processing and memory in peripheral devices becomes an increasingly attractive proposition [1, 15, 27, 33]. Placing processing power and memory capacity within a smart disk system allows functionality to be migrated from the file system into the disk (or RAID) thus providing a number of potential advantages over a traditional system. For example, when computation takes place near the data, one can ....

....blocks, directories, indirect blocks) are not exposed. Thus, research efforts have been limited to applying disksystem intelligence in a manner that is oblivious to the nature and meaning of file system traffic, e.g. improving write performance by writing blocks to the closest free space on disk [11, 33]. To fulfill their potential and retain their utility, smart disk systems must become smarter while the interface to storage remains the same. Such a system must acquire knowledge of how the file system is using it, and exploit that understanding in order to enhance functionality or increase ....

[Article contains additional citation context not shown here]

R. Wang, T. E. Anderson, and D. A. Patterson. Virtual Log-Based File Systems for a Programmable Disk. In Proceedings of the Third Symposium on Operating Systems Design and Implementation (OSDI '99), New Orleans, LA, February 1999.

....be possible in particular, deploying a substantial network just for experiments is simply not feasible. We are aware of only a few previous cases of timingaccurate storage emulation being used for performance evaluation. The most relevant example is the evaluation of eager writing by Wang et al. [25]. Under eager writing, data is written to a disk location that is close to the disk head s current location. To evaluate the benefits of having disk firmware support for eager writing, Wang et al. embedded a disk simulator in Solaris 2.6, augmented it with a RAM disk, and arranged (using the ....

....this approach, the simulator code takes the real world arrival time and the request details, and it returns the computed service time. After the appropriate real time delay, the timing loop tells the storage interface component to report completion. The emulator based evaulation of eager writing [25] used a disk simulator by Kotz et al. 15] in this way. Although it is straightforward, this first approach often does not properly handle concurrent requests. For example, a new request arrival may affect the service time of outstanding requests due to bus contention, request overlapping, or ....

[Article contains additional citation context not shown here]

Randolph Y. Wang, David A. Patterson, and Thomas E. Anderson. Virtual log based file systems for a programmable disk. Symposium on Operating Systems Design and Implementation (New Orleans, LA, 225 February 1999.

....amount of extra cost over the disk cost due to the low costs of embedded processors and memory chips. Although this new type of smart disk based architecture seems very attractive, it poses many challenges. Previous work in this area focuses on the architectural and operating system related issues [25, 34, 44]. Acharya et al. 1] on the other hand, focus on the implementation of We use the term smart disks to refer to a class of architectures that put substantial computational power on disks, such as Active Disks [1, 34, 42] and IDISKs [25] the individual database operations. Smart disks seems to ....

R. Y. Wang, T. E. Anderson, and D. A. Patterson. Virtual log based file systems for a programmable Disk. Proc. Third Symposium on Operating Systems Design and Implementation (OSDI'99), February 1999.

....very We use the term smart disks to refer to a class of architectures that put substantial computational power on disks, such as Active Disks [1] and IDISKs [25] attractive, it poses many challenges. Previous work in this area focuses on the architectural and operating system related issues [25, 34, 44]. Acharya et al. 1] on the other hand, focus on the implementation of the individual database operations. Although investigating the individual database operations on this architecture is very important, to gain more insight on the possible improvements by the smart disk architecture, we have to ....

R. Y. Wang, T. E. Anderson, D. A. Patterson. Virtual log based file systems for a programmable Disk. Proc. Third Symposium on Operating Systems Design and Implementation (OSDI'99), February 1999.

....Van Meter [27] suggested that there was general benefit in changing file systems to understand that different regions of the disk provide different bandwidths. By utilizing even more detailed disk information, several researchers have shown substantial decreases in small request response times [8, 10, 13, 46, 49]. For small writes, these systems detect the position of the head and re map data to the nearest free block in order to minimize the positioning costs [10, 46] For small reads, the SRArray [49] determines the head position when the read request is to be serviced and reads the closest of several ....

.... even more detailed disk information, several researchers have shown substantial decreases in small request response times [8, 10, 13, 46, 49] For small writes, these systems detect the position of the head and re map data to the nearest free block in order to minimize the positioning costs [10, 46]. For small reads, the SRArray [49] determines the head position when the read request is to be serviced and reads the closest of several replicas. 7 Summary This paper presents a case for track aligned extents. It demonstrates feasibility with a working prototype, and it demonstrates value with ....

Randolph Y. Wang, David A. Patterson, and Thomas E. Anderson. Virtual log based file systems for a programmable disk. Symposium on Operating Systems Design and Implementation (New Orleans, LA, 2245 February 1999.

....performance is limited by mechanical delays. In order to reduce disk I O overheads, several techniques and optimizations have been proposed. This literature includes work on optimizing the scheduling of disk requests [29, 2, 14, 11] disk arrays [7, 3, 30] and optimizing disk writes using logs [18, 10, 28]. Despite these techniques and optimizations, disk I O is still a serious problem for data intensive servers, such as Web proxies, mail and news servers, multimedia servers, and database servers. An important reason for this is that current disk controller caches are not designed to handle server ....

....There have been several works on improving disk I O performance. Most of them discuss techniques that are external to the disk drive. Typical examples are techniques for improving the prefetching and caching of disk blocks [16, 4, 24] scheduling requests [2, 11] and for optimizing disk writes [10, 18, 28]. Our techniques deal with read ahead (a form of prefetching) and caching of disk data, but their focus is on the disk controller cache. There are several important differences between the controller cache and other caches, such as the buffer cache, the Disk Caching Disk [10] or even a ....

R. Y. Wang, T. E. Anderson, and D. A. Patterson. Virtual Log Based File Systems for a Programmable Disk. In Proceedings of the 3th Symposium on Operating Systems Design and Implementation, pages 29--44, New Orleans, LA, February 1999.

....amount of extra cost over the disk cost due to the low costs of embedded processors and memory chips. Although this new type of smart disk based architecture seems very attractive, it poses many challenges. Previous work in this area focuses on the architectural and operating system related issues [25, 34, 44]. Acharya et al. 1] on the other hand, focus on the implementation of the individual database operations. Smart disks seems to be an attractive alternative especially for database applications. Therefore, investigating the individual database operations on this architecture is very important. ....

R. Y. Wang, T. E. Anderson, and D. A. Patterson. Virtual log based file systems for a programmable Disk. Proc. Third Symposium on Operating Systems Design and Implementation (OSDI'99), February 1999.

....the devices. This is because the main CPU has to go across the memory and I O buses to reach the device and incurs several hundreds of cycles for each access. In these situations, better performance can be achieved by implementing some of the functionality on the device instead of on the main CPU [15, 97, 41, 96, 79, 95, 100, 2, 99]. To implement these increasingly sophisticated functionality on devices, these devices are often equipped with a programmable processor and memory (Figure 1.1) Since the processor resides directly on the card, it incurs a much smaller overhead to access the device s resources like DMA engines ....

R. Y. Wang, T. E. Anderson, and D. A. Patterson. Virtual Log Based File Systems for a Programmable Disk. In Proceedings of the Symposium on Operating Systems Design and Implementation, New Orleans, Louisiana, February 1999.

....2, we discussed much of the work that has been done to avoid synchronous writes in FFS. As mentioned in the introduction, small writes are another performance bottleneck in FFS. Log structured file systems [Rosenblum92] are one approach to that problem. A second approach is the Virtual Log Disk [Wang99]. Log structured file systems (LFS) solve both the synchronous meta data update problem and the small write problem. Data in an LFS are coalesced and written sequentially to a segmented log. In this way, LFS avoids the seeks that a conventional file system pays in writing data back to its ....

.... and debate [Seltzer93, Blackwell95, Seltzer95, Matthews97] Building on the idea of log structured file systems, Wang and his colleagues propose an intelligent disk that performs writes at near maximum disk speed by selecting the destination of the write based upon the position of the disk head [Wang99]. The disk must then maintain a mapping of logical block number to physical location. This mapping is maintained in a virtual log which is written adjacent to the actual data being written. The proposed system exists only in simulation, but seems to offer the promise of LFS like performance for ....

Wang, R., Anderson, T., Patterson, D., "Virtual Log Based File Systems for a Programmable Disk," Proceedings of the 3rd OSDI, pp. 29--44. New Orleans, LA, Feb. 1999.

....consider solutions where computations can be moved closer to data to take advantage of these features. However, such approaches introduce many new challenges and problems that must be addressed. Previous work on this area addresses the architecture design and operating system design related issues [27,32,17]. Acharya et al. 1] on the other hand, focus on the implementation of individual database operations such as selection and join operators and of some scientific kernels on smart disk architectures. However, in order to gain further insight to the capabilities of such architectures, we need to ....

R. Y. Wang, T. E. Anderson, D. A. Patterson. Virtual log based file systems for a programmable Disk. Proc. Third Symposium on Operating Systems Design and Implementation (OSDI'99), February 1999.

.... has shown that a file system can offload the placement policy to the ondisk microprocessor and improve performance the ondisk microprocessor can determine the current disk head location and place data close to this location so as to reduce head movement and reduce the latency of write requests [13]. Partitioning file system functionality between servers and disks has fundamental implications on the policies for placement, retrieval and caching. For instance, TABLE II CHARACTERISTICS OF PDAS AND TYPICAL DESKTOPS. Characteristic PDA (Palm Pilot) Desktop Processor 16.6 MHz Motorola 68328 ....

R Y. Wang, T E. Anderson, and D A. Patterson. Virtual Log Based File Systems for a Programmable Disk. In Proceedings of third symposium on Operating systems design and implementation (OSDI), New Orleans, LA, pages 29--43, Feb 1999.

....far address the the rotational delay for writes. This is a difficult task because, if we are writing to a specific place, we have to wait for it to come under the disk head. To get around this obstacle we eagerly write to the first free block on the current track. This technique is explored in [21] to create a virtual log disk. Eager writing dramatically reduces write latency because it eliminates the seek delay and has a very small rotational delay. However it presents additional challenges because we must keep track of which blocks on each track are free and we must keep track of where we ....

....decrease read latency also increase write latency. We devote this section to modeling the reduction of write latency by eager writing. 4.5.1 Eager Write Analysis For the analysis of eager writing we assume that the blocks on a given track are uniformly distributed. This assumption is also made in [21]. Clearly this is not an exact model of what happens. For initial writes the data tends to cluster up and the behavior is similar to that of Linear Probing [18] However, as updated data blocks are written, the blocks containing the obsolete data are reallocated, leaving near random empty block. ....

Wang, R., Anderson, T., Patterson, D. Virtual Log Based File Systems for a Programmable Disk, Proceedings of the Third Symposium on Operating Systems Design and Implementation, February 1999.

....parallel computation. 164 Ongoing IDISK research is not limited to database applications. One example IDISK file system research issue is reducing write latency by having writes occur anywhere within a cylinder, leaving the decision to the IDISK processor and informing the file system later [112]. Earlier work in the DataMesh project by Wilkes, et al. foreshadowed the IDISK architecture shown in Figure 6 2, and presented a file system to exploit the DataMesh architecture [114] 115] In addition to file service, we can also imagine IDISKs offering an advantage for many other application ....

R. Wang. "Virtual log based file systems for a programmable disk," Proc. of the Third OSDI, pages 29 - 43, February 1999.

....of main memory are available today [Sun98] then the disks are necessary only for cold data and permanent storage, to protect against system failures. This means that optimizations in this area are primarily focussed on write performance and reliability. Optimizations such as immediate writes [Wang99] or non volatile memory for fast write response [Baker92] are the most helpful 1 . These types of optimization can also benefit from increased intelligence at the individual devices to implement a variety of application optimized algorithms (such as 1. but not for individual disk drives, where ....

Wang, R.Y., Anderson, T.E. and Patterson, D.A. "Virtual Log Based File Systems for a Programmable Disk" OSDI, February 1999.

....to DCD, Trail does not require NVRAM, and supports the notion of track based logging to further reduce the rotational latency. Like DCD, Trail is designed to be implemented as a self contained device driver with almost no modifications to the file system and the kernel. The virtual log paper [17] introduced the idea of eager write, which is derived from previous works [13, 12, 14] that tried to write near where the disk head is. The authors presented a device driver approach and a file system approach to exploit eager writing to improve the synchronous write performance. The performance ....

Wang, R.Y.; Anderson, T.E.; Patterson, D.A., "Virtual log based file systems for a programmable disk," Proceedings of the Third Symposium on Operating Systems Design and Implementation, p. 29-43, New Orleans, LA, USA 22-25 Feb. 1999.

.... multimedia file systems is to group the systems according to the supported multimedia data characteristics: # General file systems capable of handling multimedia data to a certain extend, e.g. FFS [Leffler et al. 90] and log structured file systems [Rosenblum 95] Seltzer et al. 93] Wang et al. 99a] # Multimedia file systems optimized for continuous multimedia data (video and audio data) e.g. SBVS [Vernick et al. 96] Mitra [Ghandeharizadeh et al. 97] CMFS [Anderson et al. 92b] PFS [Lee et al. 97] Tiger [Bolosky et al. 96] Bolosky et al. 97] Shark [Haskin 93] Tiger Shark [Haskin ....

Wang, R.Y., Anderson, T.E., Patterson, D.A.: Virtual Log Based File Systems for a Programmable Disk, Proc. of 3 rd USENIX Symp. on Operating Systems Design and Implementation (OSDI'99), New Orleans, LA, USA, February 1999, pp. 29-43 26 T. Plagemann, V. Goebel, P. Halvorsen, O. Anshus

....in host processor speed. The results for each task on a given configuration are normalized with respect to the performance of the same task on the core Active Disk configuration of the same size. 6 Related work Several researchers have recently explored the concept of programmable disks [3, 11, 15, 19, 22, 31, 38]. Initial studies on using Active Disks for databases compared the performance of a moderate size Active Disk farm ( 32 disks) with that of uniprocessor servers with conventional disks [3, 31] They demonstrated the potential of Active Disk architectures for achieving performance gains for ....

....of algorithmic research for shared nothing architectures to provide efficient implementations of a wide variety of operations. Furthermore, Active Disks can be used to perform operations for non relational data such as image processing [3, 31] and file system and security related processing [11, 15, 38, 39]. 7 Conclusions In this paper, we have evaluated Active Disk architectures for decision support databases and have compared their performance, scalability and price with that of existing scalable server architectures. In addition, we have examined the impact of individually scaling several Active ....

R. Wang, T. Anderson, and D. Patterson. Virtual log-based file systems for a programmable disk. In Proc. of OSDI'99, 1999.

No context found.

Wang, R. Y., Anderson, T. E., and Patterson, D. A. (1998b). Virtual Log Based File Systems for a Programmable Disk. Technical Report UCB/CSD 98/1031, University of California at Berkeley.

No context found.

Wang, R. Y., Anderson, T. E., and Patterson, D. A. Virtual Log Based File Systems for a Programmable Disk. In Proc. of the Third Symposium on Operating Systems Design and Implementation (New Orleans, LA, February 1999), Operating Systems Review, Special Issue, pp. 29--43.

....mirroring is more e#ective for improving a read dominant workload, a technique called eager writing is more e#ective for improving a writedominant workload. Eager writing refers to the technique of allocating a free block that is closest to the current disk head position to satisfy a write request [4, 6, 28]. Under the right conditions, by eliminating almost all of seek and rotational delay, eager writing can deliver very fast write performance without compromising reliability guarantees, even for workloads that comprise of synchronous I Os and have poor locality. What eager writing does not address, ....

....Eager writing In a traditional update in place storage system, the addresses of the incoming I O requests are mapped to fixed physical locations. In contrast, under eager writing, to satisfy a write request, the system allocates a new free block that is closest to the current disk head position [4, 6, 28]; consequently, a logical address can be mapped to di#erent physical addresses at di#erent times. A number of characteristics associated with eager writing make it suitable for transaction processing applications. The chief advantages of eagerwriting are excellent small write performance (in ....

[Article contains additional citation context not shown here]

Wang, R. Y., Anderson, T. E., and Patterson, D. A. Virtual Log Based File Systems for a Programmable Disk. In Proc. of the Third Symposium on Operating Systems Design and Implementation (New Orleans, LA, February 1999), Operating Systems Review, Special Issue, pp. 29--43.

....is more e#ective for improving a read dominant workload, a technique called eagerwriting is more e#ective for improving a write dominant workload. Eager writing refers to the technique of allocating a free block that is closest to the current disk 1 head position to satisfy a write request [4, 6, 28]. Under the right conditions, by eliminating almost all of seek and rotational delay, eager writing can deliver very fast write performance without compromising reliability guarantees, even for workloads that comprise of synchronous I Os and have poor locality. What eagerwriting does not address, ....

....Eager writing In a traditional update in place storage system, the addresses of the incoming I O requests are mapped to fixed physical locations. In contrast, under eagerwriting, to satisfy a write request, the system allocates a new free block that is closest to the current disk head position [4, 6, 28]; consequently, a logical address can be mapped to di#erent physical addresses at di#erent times. A number of characteristics associated with eagerwriting make it suitable for transaction processing applications. The chief advantages of eager writing are excellent small write performance (in ....

[Article contains additional citation context not shown here]

Wang, R. Y., Anderson, T. E., and Patterson, D. A. Virtual Log Based File Systems for a Programmable Disk. In Proc. of the Third Symposium on Operating Systems Design and Implementation (New Orleans, LA, February 1999), Operating Systems Review, Special Issue, pp. 29--43.

....the disk head 7 location and the cost of disk operations such as track switches, seeks, and rotational placement. The driver also needs information on the layout of the physical sectors on the disk. Previous proposals that depend on the knowledge of head positions have relied on hardware support [4, 27]. Unfortunately, this level of support is not always available on commodity drives. We have developed a software only head tracking method. Our scheme requires issuing read accesses to a fixed reference sector at periodic intervals. The head tracking algorithm computes the disk head position based ....

Wang, R. Y., Anderson, T. E., and Patterson, D. A. Virtual Log Based File Systems for a Programmable Disk. In Proc. of the Third Symposium on Operating Systems Design and Implementation, New Orleans, LA (February 1999), Operating Systems Review, Special Issue, pp. 29-- 43.

....the disk head location and the cost of disk operations such as track switches, seeks, and rotational placement. The driver also needs information on the layout of the physical sectors on the disk. Previous proposals that depend on the knowledge of head positions have relied on hardware support [5, 27]. Unfortunately, this level of support is not always available on commodity drives. We have 7 developed a software only head tracking method. Our scheme requires issuing read accesses to a fixed reference sector at periodic intervals. The head tracking algorithm computes the disk head position ....

Wang, R. Y., Anderson, T. E., and Patterson, D. A. Virtual Log Based File Systems for a Programmable Disk. In Proc. of the Third Symposium on Operating Systems Design and Implementation (New Orleans, LA, February 1999), Operating Systems Review, Special Issue, pp. 29--43.

....and rolls forward, VLFS traverses the virtual log backwards from the log tail towards the checkpoint. VLFS also opens up a variety of questions including how to re engineer the host disk interface and how to implement the free space compactor. These issues are discussed in a technical report [35]. 3.4 Comparing VLFS with LFS VLFS and LFS share a number of common advantages. Both can benefit from an asynchronous memory bu#er by preventing short lived data from ever reaching the disk. Both can benefit from disk reorganization during idle time. Due to eager writing, VLFS possesses a ....

Wang, R. Y., Anderson, T. E., and Patterson, D. A. Virtual Log Based File Systems for a Programmable Disk. Tech. Rep. UCB/CSD 98/1031, University of California at Berkeley, December 1998.

No context found.

Randolph Y. Wang, David A. Patterson, and Thomas E. Anderson. Virtual log based file systems for a programmable disk. Symposium on Operating Systems Design and Implementation (New Orleans, LA, 22--25 February 1999), pages 29--43. ACM, 1999.

No context found.

Randolph Y. Wang, David A. Patterson, and Thomas E. Anderson. Virtual log based file systems for a programmable disk. Symposium on Operating Systems Design and Implementation (New Orleans, LA, 22--25 February 1999.

No context found.

R. Wang, T. E. Anderson, and D. A. Patterson. Virtual Log-Based File Systems for a Programmable Disk. In OSDI '99, New Orleans, LA, February 1999.

No context found.

R. Wang, T. E. Anderson, and D. A. Patterson. Virtual Log-Based File Systems for a Programmable Disk. In OSDI '99, New Orleans, LA, February 1999.

No context found.

Randolph Y. Wang, David A. Patterson, and Thomas E. Anderson. Virtual log based file systems for a programmable disk. Symposium on Operating Systems Design and Implementation (New Orleans, LA, 22--25 February 1999.

No context found.

R. Y. Wang, T. E. Anderson, and D. A. Patterson. Virtual log based file systems for a programmable disk. In Proc. of the Third USENIX Symposium on Operating Systems Design and Implementation (OSDI'99), pages 29--43, Feb. 1999.

No context found.

R. Y. Wang, D. A. Patterson, and T. E. Anderson. Virtual log based file systems for a programmable disk. Symposium on Operating Systems Design and Implementation, pages 29--43. ACM, 1999.

No context found.

Randolph Y. Wang, Thomas E. Anderson, and David A. Patterson, "Virtual log based file systems for a programmable disk," in Proceedings of the Symposium on Operating Systems Design and Implementation (OSDI), pages 29--43, 1999.

No context found.

WANG, R. Y., ANDERSON, T. E., PATTERSON, D. A. Virtual log based file systems for a programmable disk. In Proc. Third Symposium on Operating Systems Design and Implementation (February 1999).

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