Andrew Tridgell and Paul Mackerras. The rsync algorithm. Technical Report TR-CS-96-05, Australian National University, 1997.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....set of traces. We show that there are very simple policies that achieve significant benefits over gzip compression on most web accesses, and that can be efficiently implemented at web or proxy servers. We also investigate the potential for using file synchronization techniques, such as rsync [17], for efficient web access. 1 Introduction Delta compression (or delta encoding) is a technique that encodes a target file with respect to one or several, usually similar, files called reference files. This encoding, called a delta, describes the target file in terms of the reference files, and ....

....need to choose appropriate reference pages, and this is one main focus of this paper. Delta compression vs. file synchronization: In the standard delta compression scenario, the proxy needs a copy of the old and the new version to compute a delta. File synchronization techniques such as rsync [17], on the other hand, allow the proxy to send a delta without knowing the old version in the client cache, based on only a set of hash values sent by the client as part of the request. File synchronization is essentially a special, more restricted, case of delta compression, and usually achieves ....

[Article contains additional citation context not shown here]

A. Tridgell and P. MacKerras. The rsync algorithm. Technical Report TR-CS-96-05, Australian National University, June 1996.

....such a scenario. Also, while our definition assumes a single reference file 321 , there could be several similar files that might be helpful in communicating the contents of to the client, as discussed later. In the case of the file synchronization problem, many currently known protocols [49, 16, 35] consist of a single round of communication, where the client first sends a request with a limited amount of information about t 2 to the server, and the server then sends an encoding of the current file to the client. In the case of a multi round protocol, a standard model for communication ....

....of maintaining and retrieving old versions of files as references, changes to the file at the client or a third party, or later deletion (eviction) of old versions at the server. Some typical scenarios are: Synchronization of User Files: There are a number of software packages such as rsync [49, 48], Microsoft s ActiveSync, Puma Technologies IntelliSync, or Palm s HotSync that allow synchronization between desktops, mobile devices, or web accessible user accounts. In this scenario, files or records can be updated by several different parties, and time stamps may be used to determine which ....

[Article contains additional citation context not shown here]

A. Tridgell and P. MacKerras. The rsync algorithm. Technical Report TR-CS-96-05, Australian National University, June 1996.

....outside of gossip protocols. They can be applied whenever data being reconciled can be modeled as sets and the symmetric di#erence between the sets being reconciled is small. So, for example, these protocols could be used to improve the e#ciency of file synchronization tools like Rsync [60] and Unison [8] or for reconciling replicas of distributed databases or filesystems that have diverged due to system partition [50] 1.3 Byzantine Fault Tolerance Most gossip protocol research has focused on benign failures. Chapter 4 discusses techniques for constructing gossip protocols that ....

Andrew Tridgell and Paul Mackerras. The rsync algorithm. Technical Report TR-CS-96-05, The Australian National University, Canberra 0200 ACT, Australia, 1996.

....example, distributing an updated product list to many PDAs in the same sales force. In such cases, in place reconstruction offers a lowest common denominator solution in exchange for a little lost compression. We also are developing algorithms that can perform peer to peer style delta compression [26] in an in place fashion. This allows delta compression to be used between two versions of a file stored on separate machines and is often a more natural formulation, because it does not require a computer to maintain the original version of data to employ delta compression. This works well for ....

A. Tridgell and P. Mackerras. The RSync algorithm. Technical Report TR-CS-96-05, The Australian National University, 1996.

....treat delta encoding as a content coding turned out to create extremely complex rules for handling entity tags and cache entries. We also found it difficult to define how a client could ask for delta encoding, ranges, and compression in various orders. Other proposed HTTP extensions, such as rsync[31] and cachebased compaction[4] share the definitional problems first seen with delta encoding. We can visualize the situation by depicting the HTTP message generation model as a pipeline, as shown in figure 1(a) In this figure, datatypes (bold faced terms) are transformed via processing steps ....

A. Tridgell and P. Mackerras. The rsync algorithm. Technical Report TR-CS-96-05, Dept. of Computer Science, Australian National University, June 1996.

....a T1) rather than an intermittent high bandwidth connection. Like CODA, SUNDR reduces read bandwidth with a large client site cache. In addition, SUNDR exploits commonality between files to compress data exchanged between clients and servers. The approach is similar to one used by the rsync [16] file transfer utility. 3 Design SUNDR has several design goals. The server must not see any private file data in the clear. Clients should never be tricked into believing a file contains data that wasn t written there by an authorized user. Furthermore, SUNDR should provide an easy way to ....

....after adding a debugging print statement. 59 of the writes could have saved using delta compression between the two versions. ar consisted of regenerating a 3.9MB C library after modifying one of the source files. 97 of the writes could have been saved. SUNDR can use the rsync algorithm [16] to reduce bandwidth when one file is very similar to another. However, that still leaves the problem of actually pairing up old files with new ones. In some cases this is easy. For example, dvips just truncates the old file and writes out the new one. Since allocating and freeing of blocks in ....

A. Tridgell and P. Mackerras. The rsync algorithm. Technical Report TR-CS-96-05, Australian National University, 1997.

....new version changes its name, then the patch generator will think of this as a new type, as opposed to a modified version of an existing one, and therefore will not generate any conversion routines. This di#culty of changing names arises in other areas, in particular file synchronization (e.g. TM96, BP98] etc. we should be able to apply its solutions to our patch generation system. For example, some problems could be alleviated by permitting the user to inform the generator of old # new relationships 127 old version old foo.pop: struct t int a; int b; t someTs[ int f (t ....

Andrew Tridgell and Paul Mackerras. The rsync algorithm. Technical Report TR-CS-96-05, Canberra 0200 ACT, Australia, 1996. http://samba.anu.edu. au/rsync/.

....to handle multiple data repositories at the application level and at the le system level. At the application level, some efforts have focused on using only existing system services to do peer to peer synchronization. Unfortunately, tools that use high level le metadata provided by the system[14], such as the le s name or date of last modi cation, are unreliable; they can only infer relationships between le copies from information not intended for such use. For example, if the user changes the name of one copy of a le, its relationship to other copies may be broken. Other le ....

Andrew Tridgell and Paul Mackerras, \The rsync Algorithm." Technical Report TR-CS96

....in the main repository. These intermediate versions remain private to a developer and they neither interfere with changes from others, nor clutter the version history in the main repository. Repository Replication. It is common to use CVS in combination with a replication program like rsync [19] to improve access times for developers that are physically located in different continents than the main project repository. Although certainly beneficial, this solution has the problem that conflicts arising during synchronization cannot be resolved by rsync. Instead, the synchronization fails ....

.... a deployed software system, it is more advanced in its integration between the release site (similar to SRM) and the customer site (similar to the Software Dock) In addition to these systems, others can be adapted for bug tracking (such as GNATS [7] repository synchronization (such as rsync [19]) and providing a distributed CM infrastructure (such as NUCM [21,22] or Adele [8] Basically, our vision is for a component based, fully distributed and decentralized configuration management system that manages artifacts from their incarnation to their eventual destination as an installed ....

Tridgell and P. Mackerras. The Rsync algorithm. Technical Report TR---CS---96---05, June 1996.

....Previous work has attempted to handle multiple data repositories at the application level and at the le system level. At the application level, some e orts have focused on using only existing system services. Unfortunately, le synchronization tools that use generic metadata provided by the system[14], such as the lename or date of last modi cation, are unreliable; they can only infer relationships between le copies from information not 1 intended for such use. For example, if the user changes the lename of one copy, its relationship to other copies may be broken. Other le synchronization ....

Andrew Tridgell and Paul Mackerras, \The rsync Algorithm." Technical Report TR-CS-96-05, Australian National University.

....deployment process and they provide access point to each deployment activity in the product model although their process modeling capability is far more limited than the data modeling capability. 4. 5 Delivery of Content Marimba s Castanet [15] PointCast, ZIP delivery, rdist[17] and rsync [24] implement content delivery systems. In this class of systems and technologies, the information being deployed is simply transferred from one or more information centers to a number of receiving nodes. 4.5.1 System Description Point Cast and ZIP delivery provide news multi casting services, they ....

Andrew Tridgell and Paul Mackerras. The rsync algorithm. Technical Report TR-CS-96-05, dcs-anu, anu-address, June 1996.

No context found.

Andrew Tridgell and Paul Mackerras. The rsync algorithm. Technical Report TR-CS-96-05, Australian National University, 1997.

No context found.

Andrew Tridgell and Paul Mackerras. The rsync algorithm , Technical report TR-CS-96-05, Australian National University, 1995.

No context found.

A. Tridgell and P. MacKerras. The rsync algorithm. Technical Report TR-CS-96-05, Australian National University, June 1996.

No context found.

A. Tridgell, and P. Mackerras, The rsync Algorithm. Technical Report TR-CS-96-05, Australian National University.

No context found.

A. Tridgell and P. MacKerras. The rsync algorithm. Technical Report TR-CS-96-05, Australian National University, June 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