
	Where results make sense

Topic: Shared nothing architecture

Related Topics

Single point of failure

Network

Teradata

	Oracle Launches Technology Offensive
		In shared nothing clusters, data is partitioned across multiple disks, and each server has access to a subset of those disks.
		He also argues that shared nothing is prone to single points of failure and requires bigger, more expensive hardware to scale.
		"Shared nothing is ordinarily run with dual ported disks or with the data in a storage-area network in which all disks are physically accessible from all nodes," he said.
	www.computerworld.com /printthis/2001/0,4814,61657,00.html (640 words)

	Oracle - IBM Shared Nothing Configuration - db2 convoy hashing
		IBM has traditionally used the shared nothing disk architecture for all versions of its DB2 shared database except OS/390 (see Figure 14.6).
		The shared nothing configuration utilizes isolated disk assets for each server and the database (in the IBM implementation) is hashed across all of the servers.
		The final problem with the shared nothing approach used by IBM is that to add a new cluster member the data must be re-hashed across all members requiring database down time.
	www.dba-oracle.com /t_ault_125_db2_convoy_hashing.htm (356 words)

	EDM/2 - OS/2 Routes - Shared Nothing
		Shared Nothing is a technology that IBM is counting on to take us into the future.
		I will share what I know about this very compelling technology, as well as engage in some outright speculation regarding the future implementation of this form of parallel processing.
		From this we can speculate that the applications which will perform best in a shared nothing multiprocessing environment will be those which make heavy use of general purpose worker threads.
	www.edm2.com /0409/os2routes.html (1431 words)

	Greenplum - Bizgres MPP - Shared Nothing (Site not responding. Last check: 2007-10-23)
		Shared-everything architectures, in which processors share a single large disk and memory, are well suited to OLTP workloads.
		However, shared-everything architectures are quickly overwhelmed by the full table scans, multiple complex table joins, sorting, and aggregation operations against vast volumes of data that represent the lion’s share of BI workloads.
		Bizgres MPP's shared nothing architecture, which leverages a number of self-contained, parallel processing units, is a proven and effective solution to support large-scale data warehousing demands.
	www.greenplum.com /products/sharedNothing.php (256 words)

	Real Application Cluster - a Whatis.com definition
		According to Oracle, RAC's shared disk method of clustering databases: increases scalability because servers can easily be added or subtracted to meet current needs, lowers costs because companies don't have to buy high-end servers, and improves availability because if one server fails, another can assume its workload.
		RAC's shared disk architecture is an unusual approach to database clustering.
		Shared nothing architecture partitions data and only gives each server access to its own disk subsystem, while shared disk architecture gives all servers access to the entire database.
	whatis.techtarget.com /gDefinition/0,294236,sid4_gci871791,00.html (283 words)

	Shariq Rizvi - Assignment 0
		Shared disk: Each processor has its own memory but a shared set of storage disks is used.
		Shared nothing: These involve multiple machines that share no hardware but communicate with each other using a high-speed LAN.
		Secondly, using a shared nothing architecture allows the system storage to scale with the number of processors too - thus increasing the I/O bandwidth of the system, without the use of specialized disk controllers.
	www.cs.berkeley.edu /~ejr/GSI/cs267-s04/homework-0/results/rizvi (678 words)

	High-Performance ETL
		DB2 UDB ESE uses a "shared-nothing" approach, which means that no data and no memory are shared among database partitions.
		This architecture is unique compared to other relational database management systems.
		If you don't take DB2's architecture into account at the design stage, ETL processes may slow to a crawl no matter how powerful your systems are or how much memory you add.
	www.db2mag.com /shared/printableArticle.jhtml?articleID=173600267 (1685 words)

	Friendster goes PHP (Site not responding. Last check: 2007-10-23)
		The absolute fastest performance because nothing is going to beat decent C code and the frontend flexibility to quickly modify the user experience and launch new features through PHP's very easy-to-learn and friendly frontend language.
		The design and architecture of the system means more than the language it was written in, and you can write a kickass app or a shitty app in Java or PHP.
		After her termination, with nothing to more to lose, she didn't badmouth the company, and actually had some nice things to say about some of the people she worked with.
	troutgirl.com /blog/index.php?/archives/22_Friendster_goes_PHP.html (13288 words)

	RepDB - Preemptive Replication for Databases (Site not responding. Last check: 2007-10-23)
		Similar to multiprocessors, various cluster system architectures are possible: shared-disk, shared-cache and shared-nothing.
		Shared-disk and shared-cache require a special interconnect that provide a shared space to all nodes with provision for cache coherence using either hardware or software.
		Shared-nothing (or distributed memory) is the only architecture that supports our autonomy requirements without the additional cost of a special interconnect.
	www.sciences.univ-nantes.fr /lina/ATLAS/RepDB/presentation.html (872 words)

	5 Shared-Nothing Systems (Site not responding. Last check: 2007-10-23)
		The shared-nothing (SN) architecture consists of multiple nodes with private memory, disks and input/output devices, connected via interconnection network.
		For large-scale computing, SN architecture is now the most popular solution.
		It introduces architecture with enormous complexity, resulting in a new challenge for system designers.
	homepages.cwi.nl /~marcin/thesis/node17.html (226 words)

	DB2 UDB Information Zone (Site not responding. Last check: 2007-10-23)
		Many years ago before the availability of systems with multiple processors, gigabytes of memory, and storage area networks it was easy to identify a shared-nothing architecture because the hardware that supported the DPF was not shared.
		None-the-less the DPF still implements a shared-nothing architecture simply because each partitioned database server, logical or physical, manages one and only one partition of the database.
		The shared nothing architecture of the Database Partitioning Feature (DPF) reduces contention for system resources and provides near linear scalability by dividing very large databases into smaller part (partitions) that are individually managed by each partitioned database server.
	www.tendigit.com /izone/briefs/dpf.html (755 words)

	epowiki: Share Nothing Architecture
		In a shared nothing architecture no caches are kept.
		All data are kept in a database and shared via the database.
		This obviously puts a lot of stress on the database, but the thought is the database can be made to scale more readily than applications.
	www.possibility.com /epowiki/Wiki.jsp?page=ShareNothingArchitecture (176 words)

	Parallel R-tree Spatial Join for a Shared-Nothing Architecture (ResearchIndex) (Site not responding. Last check: 2007-10-23)
		This paper proposes a Replicated Parallel Packed R-tree and its use in performing the parallel R-tree join.
		We examine performance using the Digital Chart of the World Data on a shared nothing machine.
		0.3: Master-Client R-trees: A New Parallel R-tree Architecture - Schnitzer, Leutenegger (1998)
	citeseer.ist.psu.edu /286183.html (297 words)

	[No title] (Site not responding. Last check: 2007-10-23)
		Shared-everything architecture provides better performance than the shared-nothing architecture but it is not scalable to large system sizes.
		Hierarchical architectures have been proposed to incorporate the best features of these two architectures (i.e., the shared-nothing and the shared-everything).
		In this paper, we present a detailed simulation study comparing the performance of the two-level hierarchical architecture with that of the shared-nothing and shared-everything architectures.
	www.scs.carleton.ca /research/tech_reports/1997/TR-97-02.html (187 words)

	Do You PHP? by Rasmus Lerdorf (Site not responding. Last check: 2007-10-23)
		That doesn't mean I don't believe in structure and frameworks, but I do believe in people having the power to come up with their own to match their environment.
		The shared-nothing architecture of PHP where each request is completely distinct and separate from any other request leads to infinite horizontal scalability in the language itself.
		There is nothing wrong with direct database calls' making use of all the tricks and cheats your chosen database has to offer, to tweak as much performance as possible out of it.
	www.oracle.com /technology/pub/articles/php_experts/rasmus_php.html (2809 words)

	IEEE Transactions on Parallel and Distributed Systems
		In the shared nothing (SN) architecture, neither disks nor memory is shared.
		In the shared disk (SD) architecture, all disks are accessible to all nodes while in the shared intermediate memory (SIM) architecture, a shared intermediate level of memory is introduced.
		We describe a binding environment for the AND and OR parallel execution of logicprograms that is suitable for both shared and nonshared memory multiprocessors.
	csdl2.computer.org /comp/trans/td/1994/02/l2toc.xml (1665 words)

	Entry-Level Solutions: Sun, IBM and Infospace
		DB2 UDB EEE employs a shared-nothing software architecture, which enables it to efficiently support very large databases by dividing the database into partitions; each which can be stored and managed on any combination of multiple logical (partitioning on the same machine) or physical (partitioning on separate machines) nodes of a shared-nothing hardware platform.
		For example, in the case where two system nodes are paired together, and each pair has access to shared disks (disks connected to two nodes) if one of the nodes in a pair fails, the other nodes can take over and the system continues to operate (an active-active failover scenario).
		While this method provides quick takeover of a failed node, there may be an impact on performance due to an increased load on the takeover node.
	www.sun.com /servers/entry/solutions/ibm/ibm-db2.html (4787 words)

	Parallel Sorting on a Shared-Nothing Architecture using Probabilistic Splitting - DeWitt, Naughton, Schneider ... (Site not responding. Last check: 2007-10-23)
		Parallel Sorting on a Shared-Nothing Architecture using Probabilistic Splitting - DeWitt, Naughton, Schneider (ResearchIndex)
		Parallel Sorting on a Shared-Nothing Architecture using Probabilistic Splitting (1992)
		D.J.DeWitt, D.J. Naughton and D.A.Schneider "Parallel Sorting on a Shared-Nothing Architecture Using Probabilistic Splitting" Proceedings First International Conference on Parallel and Distributed Info Systems, IEEE Press, Jan 1992.
	citeseer.ist.psu.edu /399696.html (313 words)

	High-End Data Warehouses - "Shared Nothing" RDBMSs (Site not responding. Last check: 2007-10-23)
		In this Perspective, Gartner Dataquest discusses the market for RDBMSs targeting high-end data-warehousing, that is, those with "shared nothing" parallel architectures.
		The license revenue for RDBMS software in this specialized segment is less than 1 percent of the license revenue in the broader market of DBMSs for decision support.
		The terabytes of data generated by e-commerce will increase the demand for "shared nothing" architecture RDBMSs for the data warehouse.
	ncr.com /repository/research_reports/data_warehousing/high_end_dw.htm (85 words)

	Feld Thoughts
		Tim Wolters – an extremely capable CTO – has an introduction to how he is approaching this at Collective Intellect.
		He’s taking a page from Google’s playbook and developing a web service based on a “shared nothing architecture.” On Friday, I had two different discussions about scalable architectures (e.g.
		I expect we’ll hear a lot more about this in 2006 as a small percentage of the flood of web apps created in 2005 become popular enough to have real scale issues.
	www.feld.com /blog/archives/2005/12/shared_nothing.html (598 words)

	Tuning IBM DB2* Databases for Intel® Itanium® 2-Based Systems
		A shared-nothing architecture allows a workload to be divided into multiple logical units or partitions, and each logical unit is allocated with its own computing resources, such as processors, memory, and disks.
		The shared-nothing architecture can be applied not only to multiple physical nodes or to a cluster of machines, but also to a single machine with multiple processors or an SMP machine.
		In this benchmark, we divide the TPC-H workload among the four Itanium 2 processors so that each processor works as independently as possible from the other processors in the system.
	www.intel.com /cd/ids/developer/asmo-na/eng/76370.htm?page=4 (1724 words)

	PACT-95 Article: E. Ozkarahan. Hardware and Software Platform for Information Processing
		In view of the limitations of the classical uniprocessor architecture and of the characteristics of contemporary information processing which may involve multi-media objects, a hardware software platform is defined over parallel processing.
		Such a platform possesses a system architecture which combines parallel algorithms, parallel data structures, parallel processors, and an efficient input output architecture.
		Our research that uses this platform in the implementation of a multi-media information system and its methodology are outlined.
	www.pact.sscc.ru /conference/pact95/articles/6660964042.html (149 words)

	VLDB 1996: 366-377
		The first implementation step is a translation into a specially designed algebra for flat sequences, having only flat parallelism: the translation ``flattens'' the nested parallelism, and we prove that it preserves the asymptotic parallel complexity.
		The second step consists in an implementation of the sequence algebra on a shared nothing architecture.
		To account for the communication costs we chose as shared nothing model the recently proposed LogP model.
	www.vldb.org /dblp/db/conf/vldb/Suciu96a.html (721 words)

	Jeff Clavier's Software Only: Thoughts on Shared Nothing Architectures
		Buddy Brad Feld has a great post on Shared Nothing Architecture, as a potential solution to performance and reliability issues faced by services I use on a day to day basis: TypePad and del.icio.us (and to some extent, Bloglines - though I don't use it so much now).
		I had actually spotted that del.icio.us was down as well , and was about to write my own piece out of frustration, but Brad is summarizing the situation well.
		The ultimate Shared Nothing Architecture relies on mirrored data centers in different physical geographies that allows a system to switch over in (quasi) real-time in case of any type of failure (power, hardware, database, etc.) - and this is expensive to deploy.
	blog.softtechvc.com /2005/12/thoughts_on_sha.html (728 words)

	Scaling Up vs. Scaling Out (Site not responding. Last check: 2007-10-23)
		Essentially, shared-nothing architecture means that each system operates independently.
		Although scaling out provides great answers to the inherent limitations in scale-up architecture, this method is no stranger to Murphy's Law, either.
		At this point in the technology lifecycle, scaling out requires increased management overhead that is potentially as great as the performance gains it offers.
	www.sqlmag.com /articles/index.cfm?articleid=8755 (680 words)

	, from - White Papers, Webcasts and Case Studies - TechRepublic
		Webcast: The New Exchange Architecture: Using Continuity to Lower Cost and Improve Email Availability
		In addition, you allow us to share your information with companies that produce products or services featured in the library--so that such companies may contact you with information and offers regarding their products and services.
		This enables us to keep the library a free service.
	whitepapers.techrepublic.com.com /abstract.aspx?...&subj=&promo=100112 (376 words)

Try your search on: Qwika (all wikis)