Search Results

Search found 23827 results on 954 pages for 'software architecture'.

Page 474/954 | < Previous Page | 470 471 472 473 474 475 476 477 478 479 480 481  | Next Page >

• Deployment Setup (.Net) - Search target machine -> Registry search (64 bit)

  - by Joonas Kirsebom

  I have a windows installer project which installs some software (winform, service, mce addin). During the installation I need to search the machine for a registry key. This is done with with the "Launch Condition" - "Add Registry Search" (Deployment Project). I have filled out all the properties right, and checked against the regestry that the value actually can be found. The problem is that the "Registry Search" searches in the x86 part of the registry (HKEY_LOCAL_MACHINE\SOFTWARE\Wow6432Node\...) although my system is a x64 and the deployment setup is also set to x64. Does anyone know how to force the "Registry Search" to search the x64 registry? Or know about a workaround? The weird thing about this, is that Registry setting in the deployment setup is writing to the right registry (x64). My idea is that the "Registry Search" program is only developed to the x86 architecture, and therefore can't read the right registry. I found this article from microsoft, so it seams that they know about this problem. https://connect.microsoft.com/VisualStudio/feedback/ViewFeedback.aspx?FeedbackID=110105&wa=wsignin1.0#details My system is: Windows 7 64bit Visual Studio 2008

  Read the article

• monitoring TCP/IP performance on Solaris

  - by Andy Faibishenko

  I am trying to tune a high message traffic system running on Solaris. The architecture is a large number (600) of clients which connect via TCP to a big Solaris server and then send/receive relatively small messages (.5 to 1K payload) at high rates. The goal is to minimize the latency of each message processed. I suspect that the TCP stack of the server is getting overwhelmed by all the traffic. What are some commands/metrics that I can use to confirm this, and in case this is true, what is the best way to alleviate this bottleneck?

  Read the article

• Stack Overflow when debugging application in iPhone simulator

  - by mjdth

  I'm getting this every time I attempt to debug my app in the simulator: [Session started at 2010-05-11 16:16:52 -0500.] GNU gdb 6.3.50-20050815 (Apple version gdb-1467) (Wed Apr 21 06:57:21 UTC 2010) Copyright 2004 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type "show copying" to see the conditions. There is absolutely no warranty for GDB. Type "show warranty" for details. This GDB was configured as "x86_64-apple-darwin".sharedlibrary apply-load-rules all Attaching to process 51573. Program received signal: “EXC_BAD_ACCESS”. Data Formatters temporarily unavailable, will re-try after a 'continue'. (Cannot call into the loader at present, it is locked.) I've looked around and found a few similar cases, but they all seem to be related to a missing file and an extra necessary build phase. I'm getting no notification of a missing file here so I'm not sure where to start to fix this and get the app running again. Thanks for any insight!

  Read the article

• C# COM Cross Thread problem

  - by user364676

  Hi, we're developing a software to control a scientific measuring device. it provides a COM-Interface defines serveral functions to set measurement parameters and fires an event when it measured data. in order to test our software, i'm implementing a simulation of that device. the com-object runs a loop which periodically fires the event. another loop in the client app should now setup up the com-simulator using the given functions. i created a class for measuring parameters which will be instanciated when setting up a new measurement. // COM-Object public class MeasurementParams { public double Param1; public double Param2; } public class COM_Sim : ICOMDevice { public MeasurementParams newMeasurement; IClient client; public int NewMeasurement() { newMeasurment = new MeasurementParam(); } public int SetParam1(double val) { // why is newMeasurement null when method is called from client loop newMeasurement.Param1 = val; } void loop() { while(true) { // fire event client.HandleEvent; } } } public class Client : IClient { ICOMDevice server; public int HandleEvent() { // handle this event server.NewMeasurement(); server.SetParam1(0.0); } void loop() { while(true) { // do some stuff... server.NewMeasurement(); server.SetParam1(0.0); } } } both of the loops run in independent threads. when server.NewMeasurement() is called, the object on the server is set to a new instance. but in the next function, the object is null again. do the same when handling the server-event, it works perfectly, because the method runs in the servers thread. how to make it work from client-thread as well. as the client is meant to be working with the real device, i cannot modify the interfaces given by the manufactor. also i need to setup measurements independent from the event-handler, which will be fired not regulary. i assume this problem related to multithreaded-COM behavior but i found nothing on this topic.

  Read the article

• I have two choices of Master's classes this fall. Which is the most useful?

  - by ahplummer

  (For background purposes and context): I am a Software Engineer, and manage other Software Engineers currently. I kind of wear two hats right now: one of a programmer, and one as a 'team lead'. In this regard, I've started going back to school to get my Master's degree with an emphasis in Computer Science. I already have a Bachelor's in Computer Science, and have been working in the field for about 13 years. Our primary development environment is a Windows environment, writing in .NET, Delphi, and SQL Server. Choice #1: CST 798 DATA VISUALIZATION Course Description: Basically, this is a course on the "Processing" language: http://processing.org/ Choice #2: CST 711 INFORMATICS Course Description: (From catalog): Informatics is the science of the use and processing of data, information, and knowledge. This course covers a variety of applied issues from information technology, information management at a variety of levels, ranging from simple data entry, to the creation, design and implementation of new information systems, to the development of models. Topics include basic information representation, processing, searching, and organization, evaluation and analysis of information, Internet-based information access tools, ethics and economics of information sharing.

  Read the article

• Development/runtime Licensing mechanism for a C# class library?

  - by Darryl

  I'm developing a .Net class library (a data provider) and I'm starting to think about how I would handle licensing the library to prospective purchasers. By licensing, I mean the mechanics of trying to prevent my library from being used by those who haven't purchased it, not the software license (i.e., Apache, Gnu, etc). I've never dealt with licensing, and in the past, I've always developed apps, not libraries. I don't want to make things difficult for my customers; know it is not possible to make it ironclad. Just some mechanism that gives me decent protection without making the customer jump through hoops or gnash their teeth. I think the mechanism would check for a valid license when the class is being used in development mode, and not in runtime mode (when the customer's software is released to their customers). I think libraries are typically sold per developer, but I'm not sure how that could be accomplished without making the mechanism odious for my customers; maybe that gets left to the honor system. I Googled this and found many approaches. Ideally, I'd like to do something that is generally accepted and common, the "right" way class libraries are licensed, if that exists, rather than making my customers deal with yet another license mechanism. A firm push in the right direction will be greatly appreciated!

  Read the article

• When connecting SAP Business One to SQL Server 2005, what is the

  - by Nick

  we have SAP Business One - Fourth Shift Edition running here at a small manufacturing company. The consulting company that has come in to do the installation/implementation uses the "sa" id/pass to initially connect to the database to get the list of companies. From then on, I have to assume that its the sa id/pass that is being used to connect the client software to the database. Is this appropriate? I dont know where this data is being stored... as an ODBC connection? directly in the registry somewhere? Is it secure? Would it be better to set the users network ID in the database security and then use the "trusted connection" setting instead? Or do most people create a separate login in the database for each user and use that in the client settings? seems like the easiest way would be to add the users network login to the sql server security so they can use the "trusted connection"... but then wouldn't that allow ANY software to connect to the database from that machine? So anyways: what are the best-practices for setting this up?

  Read the article

• Suscribers locking during snapshot replication

  - by remi_bourgarel

  Hi all :) Here is my architecture : I have a main server and 4 replica (the servers are synchronized with snapshot replication). The replication is working fine, except for one thing : during the replication a lot of SELECT request on one of the replica fail with a time-out. Here are my questions : Can I avoid these time-out ? If I can't : how can I detect the start and the end of the replication to redirect all the request on one of the replica to the main ? Thanks Sorry if you already answered to that kind of question but I couldn't find anything.

  Read the article

• Installing Java 1.5 on Ubuntu?

  - by StackedCrooked

  I already have Java 1.6, but I need to test something with 1.5. I have downloaded the .bin file from http://java.sun.com/javase/downloads/index_jdk5.jsp using the Sun Download Manager. Now I want to create a deb file from this bin file: $ fakeroot make-jpkg java_ee_sdk-5_01-linux.bin Creating temporary directory: /tmp/make-jpkg.Zpm1Y7LbZ0 Loading plugins: blackdown-j2re.sh blackdown-j2sdk.sh common.sh ibm-j2re.sh ibm-j2sdk.sh j2re.sh j2sdk-doc.sh j2sdk.sh j2se.sh sun-j2re.sh sun-j2sdk-doc.sh sun-j2sdk.sh Detected Debian build architecture: i386 Detected Debian GNU type: i486-linux-gnu No matching plugin was found. Removing temporary directory: done How can I fix the "No matching plugin was found." error?

  Read the article

• Script to copy files on CD and not on hard disk to a new directory

  - by John22

  I need to copy files from a set of CDs that have a lot of duplicate content, with each other, and with what's already on my hard disk. The file names of identical files are not the same, and are in sub-directories of different names. I want to copy non-duplicate files from the CD into a new directory on the hard disk. I don't care about the sub-directories - I will sort it out later - I just want the unique files. I can't find software to do that - see my post at SuperUser http://superuser.com/questions/129944/software-to-copy-non-duplicate-files-from-cd-dvd Someone at SuperUser suggested I write a script using GNU's "find" and the Win32 version of some checksum tools. I glanced at that, and have not done anything like that before. I'm hoping something exists that I can modify. I found a good program to delete duplicates, Duplicate Cleaner (it compares checksums), but it won't help me here, as I'd have to copy all the CDs to disk, and each is probably about 80% duplicates, and I don't have room to do that - I'd have to cycle through a few at a time copying everything, then turning around and deleting 80% of it, working the hard drive a lot. Thanks for any help.

  Read the article

• Setting minimum size limit for a window in java swing

  - by shadyabhi

  I have a JFrame which has 3 JPanels in GridBagLayout.. Now, when I minimize a windows, after a certain limit, the third JPanel tends to disappear. I tried setting minimizing size of JFrame using setMinimumSize(new Dimension(int,int)) but no success. The windows can still be minimized. So, I actually want to make a threshhold, that my window cannot be minimized after a certain limit. How can I do so? Code:- import java.awt.Dimension; import javax.swing.JFrame; public class JFrameExample { public static void main(String[] args) { JFrame frame = new JFrame("Hello World"); frame.setDefaultCloseOperation(JFrame.EXIT_ON_CLOSE); frame.setMinimumSize(new Dimension(400, 400)); frame.setVisible(true); } } Also: shadyabhi@shadyabhi-desktop:~/java$ java --showversion java version "1.5.0" gij (GNU libgcj) version 4.4.1 Copyright (C) 2007 Free Software Foundation, Inc. This is free software; see the source for copying conditions. There is NO warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. Usage: gij [OPTION] ... CLASS [ARGS] ... to invoke CLASS.main, or gij -jar [OPTION] ... JARFILE [ARGS] ... to execute a jar file Try `gij --help' for more information. shadyabhi@shadyabhi-desktop:~/java$ Gives me output like

  Read the article

• How can I write System preferences with Java? Can I invoke UAC?

  - by Jonas

  How can I write system preferences with Java, using Preferences.systemRoot()? I tried with: Preferences preferences = Preferences.systemRoot(); preferences.put("/myapplication/databasepath", pathToDatabase); But I got this error message: 2010-maj-29 19:02:50 java.util.prefs.WindowsPreferences openKey VARNING: Could not open windows registry node Software\JavaSoft\Prefs at root 0x80000002. Windows RegOpenKey(...) returned error code 5. Exception in thread "AWT-EventQueue-0" java.lang.SecurityException: Could not open windows registry node Software\JavaSoft\Prefs at root 0x80000002: Access denied at java.util.prefs.WindowsPreferences.openKey(Unknown Source) at java.util.prefs.WindowsPreferences.openKey(Unknown Source) at java.util.prefs.WindowsPreferences.openKey(Unknown Source) at java.util.prefs.WindowsPreferences.putSpi(Unknown Source) at java.util.prefs.AbstractPreferences.put(Unknown Source) at biz.accountia.pos.install.Setup$2.actionPerformed(Setup.java:43) I would like to do this, because I want to install an embedded JavaDB database, and let multiple users on the computer to use the same database with the application. How to solve this? Can I invoke UAC and do this as Administrator from Java? And if I log in as Administrator when writing, can I read the values with my Java application if I'm logged in as a User?

  Read the article

• catch-22 with apt-get

  - by Mark J Seger

  I'd recently installed a package and discovered my scp stopped working. After removing and installing some things I got it fixed but then I stated getting errors in apt-get about dpkg: error: configuration error: /etc/dpkg/dpkg.cfg.d/multiarch:1: unknown option 'foreign-architecture' so I commented it out and thought that fixed the problem until I discovered the chrome icon in my launcher turned into a ? and chrome no longer worked. I tried to reinstall it and got the apt-get error: "ambiguous package name 'libglib2.0-0' with more than one installed instance" if I try to remove libglib2 I get the error The following packages have unmet dependencies: iceape-browser : Depends: iceape but it is not going to be installed iceape-chatzilla : Depends: iceape (>= 2.7.11) but it is not going to be installed Depends: iceape (<= 2.7.11-1.1~) but it is not going to be installed E: Error, pkgProblemResolver::Resolve generated breaks, this may be caused by held packages. so then I tried to remove icape-browser and it complained about the 2 instances of libglinb2. in fact virtually any command I issue does the same thing so I don't know what I can do to untangle things.

  Read the article

• Disk / system configuration for log collection / syslog server

  - by Konrads

  I am looking into building a syslog / logging infrastructure and am pondering about some architecture best practices. Essentially, I see that a syslog system needs to support two conflicting workloads: log collection. Potentially massive streams of data need to be written quickly to disks and indexed. log querying. logs will be queried by both fixed fields such as date and source as well as text search. What is the best disk/system setup assuming I'd like to keep it to a single server for now? Should I use SSDs or ramdisk to off-load some processing? some disks in stripe and some in raid5? I am particularly eyeing Graylog2 with ElasticSearch/MongoDB

  Read the article

• how to do these in visio 2007?

  - by user285825

  I am very annoyed with this software. I am unable to do many things with this software. In the book 'UML distilled' many features of UML are discussed which I am not sure how to accomplish with visio 2007. For instance, 1) I can't find the unary association. In the UML static structure under shapes panel on left, there are a lot of components like package, class, blah, blah, binary association, blah, blah, association class. But where is unary association. 2) For sequence diagram, I created a message (any of sync, async, call type message, ordinary message). The I tried to incorporate some parameter information. I went to the properties. There were a category called arguments. But selecting that shows a table where arguments are supposed to be shown. But all are diasbled. 3) For sequence diagram, there is no component delete (the big X) in the UML Sequence. 4) For class diagram, there is supposed to be a comment compartment where comments like mentioning the responsibilities are allowed using a comment prefixed with "--". But I am not sure how to accomplish them. 5) there is also supposed to be a way to indicate static properties of a class. But I am not sure how to do that in visio. 6) there is supposed to be a stereotype for class <. But in the stereotype drop down there is no stereotype. 7) where is the ball and socket component?

  Read the article

• Getting Recognition for Open-Source Computer Language Projects

  - by Jon Purdy

  I like language a lot, so I write a lot of language-based solutions for programming, automation, and data definition. I'm very much a believer in open-source software, so lately I've started to push these projects to Sourceforge when I start them. I feel that these tools could be quite valuable in the right hands, and that they fill niches that otherwise go unfilled. The trouble, for me, is gaining recognition. No matter how useful the software I write, after a certain point I can no longer come up with anything to add or improve. Basically no one but me uses it, so it's not being attacked from enough angles to discover any new weaknesses. I cannot work on a project that doesn't have anything to do, but I won't have anything to do unless I gain recognition by working on it! This is greatly discouraging. It's like giving what you think is a really thoughtful gift to someone who just isn't paying attention. So I'm looking for advice on how to network and disseminate information about my projects so that they don't fizzle out like this. Are there any sites, newsgroups, or mailing lists that I've been completely missing?

  Read the article

• What is "2LUN" mode in connection with RAID?

  - by naxa

  I've came across RAID products that also list JBOD (just a bunch of disks) mode and 2LUN mode. What the heck is 2LUN mode? I could not find a description; the closest thing seems to be LUN 'logical unit number' but I don't get the 2LUN thing. UPDATE 1 This is what Wikipedia has to say about JBOD: JBOD (derived from "just a bunch of disks"): an architecture involving multiple hard drives, while making them accessible either as independent hard drives, or as a combined (spanned) single logical volume with no actual RAID functionality. So JBOD can actually mean two different (albeit related) things. Answer of Guest says 2LUN means no spanning. Does this suggest that 2LUN would simply mean the JBOD-variant with no span?

  Read the article

• Struggling to create correct relationships in MS Access

  - by Yandawl

  http://img714.imageshack.us/img714/7820/croppercapture1.png Basically: an award(course) has many units, which can be either optional or core(mandatory), depending on the award. So for example: the unit 'Advanced Software Engineering' maybe a core unit for the award 'Software Engineering BSc' but only an optional unit for the course 'Web Technology BSc'. I've used flags for that purpose. A student is enrolled on an award so I need to get a complete list of core and optional units (bearing in mind that a student chooses 1 out of many possible optional units). Also, these units have events, e.g, a lecture, workshop or seminar, etc. and those events have sessions or instances of events where students enrolled on that particular unit are required to attend, and those attendances are stored in a separate table to form a register. So I need a hierarchy of expanding the tables something like this I guess: Awards - Students - Units - Sessions - Attendances Any help with this would be appreciated... It's blowing my mind and I'm really close to going insane! My tutor didn't spot I'd got it wrong when I showed my original data model to him and it's due in next week! Thank you :D

  Read the article

• Who architected / designed C++'s IOStreams, and would it still be considered well-designed by today'

  - by stakx

  First off, it may seem that I'm asking for subjective opinions, but that's not what I'm after. I'd love to hear some well-grounded arguments on this topic. In the hope of getting some insight into how a modern streams / serialization framework ought to be designed, I recently got myself a copy of the book Standard C++ IOStreams and Locales by Angelika Langer and Klaus Kreft. I figured that if IOStreams wasn't well-designed, it wouldn't have made it into the C++ standard library in the first place. After having read various parts of this book, I am starting to have doubts if IOStreams can compare to e.g. the STL from an overall architectural point-of-view. Read e.g. this interview with Alexander Stepanov (the STL's "inventor") to learn about some design decisions that went into the STL. What surprises me in particular: It seems to be unknown who was responsible for IOStreams' overall design (I'd love to read some background information about this — does anyone know good resources?); Once you delve beneath the immediate surface of IOStreams, e.g. if you want to extend IOStreams with your own classes, you get to an interface with fairly cryptic and confusing member function names, e.g. getloc/imbue, uflow/underflow, snextc/sbumpc/sgetc/sgetn, pbase/pptr/epptr (and there's probably even worse examples). This makes it so much harder to understand the overall design and how the single parts co-operate. Even the book I mentioned above doesn't help that much (IMHO). Thus my question: If you had to judge by today's software engineering standards (if there actually is any general agreement on these), would C++'s IOStreams still be considered well-designed? (I wouldn't want to improve my software design skills from something that's generally considered outdated.)

  Read the article

• Java Framework Choice Question.

  - by Sarang

  We do have many frameworks available in Java. Struts, Swing, JSF 2.0, Spring etc are used as per their priority. Actually, I don't know how many they are as well! But, as I am fresher to Java, even learning after their architecture, I cannot decide which framework can be used with what type of Projects ! Also, I am confused with mixed use of framework like Spring + JSF. What is the benefit to it ? Another thing making me confusing is about the UI components available in market. Like, we do have Primefaces, Ice-faces, MyFaces, Rich-faces. They may or may not have been supporting AJAX in-built. They may contains some bugs as well. What is best choice for Framework + UI component that can directly provide a best feet solution for any project ?

  Read the article

• Splitting build cross the network?

  - by Dandikas

  Is there a known solution for splitting build process cross the network machines? Use case: We are an average software development company. We own around 50 development workstations (Quad Core 2.66Ghz, 4 GB ram, 200 GB raid). No need to tell that at any single moment not every machine is loaded to the max. There are 5 to 15 projects running simultaneously at any single moment. Obviously all of them are continuously build on server, than deployed to proper environment. Single project build is taking from 3 to 15 minutes. The problem: Whenever we build 5 projects in a row the last project is going to be ready after around 25 - 50 minutes. Building in parallel does not solve the problem (build is only a part of the game, than you need to deploy, run tests etc.) YES the correct solution is to add another build server, but "That involves buying new Expensive hardware, and we already spent a lot!". Yea, right(damn them)! Anyway. What about splitting build among developers workstation? Lets say whenever we need to build project "A" we check 5 workstations and start build on all that are not overloaded. The build can be canceled by a developer if he really needs all the power of his machine as long as there is at least 1 machine that is still building. After build is finished deployment can be performed to a proper environment (hosted on some server, not on workstation :) ). The bigger the company the more this makes sense to me. Anyone tried something like this? Are there any good practices? Any helpful software? (90% of the projects are .net C#)

  Read the article

• Replacing accented/umlauted characters with their unadorned counterparts in C# [closed]

  - by Andrew Rollings

  Duplicate of 249087 I have a bunch of user generated addresses that may contain characters with diacritic marks. What is the most effective (i.e. generic) way (apart from a straightforward replace) to automatically convert any such characters to their closest English equivalent? E.g. any of àâãäå would become a æ would become the two separate letters ae ç would become c any of èéêë would become e etc. for all possible letter variations (preferably without having to find and encode lookups for each diacritic form of the letter). (Note: I have to pass these addresses on to third party software that is incapable of printing anything other than English characters. I'd rather the software was capable of handling them, but I have no control over that.) EDIT: Never mind... Found the answer [here][2]. It showed up in the "Related" section to the right of the question after I posted, but not in my prior search or as a pre-post suggestion. Hmm. I added the 'diacritics' tag to the other question in any case. EDIT 2: Jeez! Who voted this -1 after I closed it?

  Read the article

• How should a programmer go about getting started with Flash/Flex/ActionScript?

  - by Graphics Noob

  What is the shortest path between zero (ie no flash related development software on my computer or information about where to obtain it or get started) to running a "hello world" ActionScript? I'm hoping for an answer that gives step by step instructions about exactly what software is needed to get started, an example of some "hello world" code, and instructions for compiling and running the code. I've spent more time than I think should be necessary researching this question and not found much information. Hopefully this question will be found by programmers like me who want to get started with Flash/Flex/ActionScript (After my morning of researching I still don't even know what terminology to use so I'll just throw it all out there). ActionScript tutorials I've found are focused on programming concepts, ie logic, branching, OOP, etc, and some even have code examples to download, but not a single one I've found explains how to compile and run the code. They all seem to assume you have an IDE standing by but no knowledge of programming, exactly the opposite of the position I'm in. Here are the most related SO questions I've found: http://stackoverflow.com/questions/59083/what-is-adobe-flex-is-it-just-flash-ii http://stackoverflow.com/questions/554899/getting-started-with-flex-3-ui-actionscript-programming http://stackoverflow.com/questions/2123105/how-to-learn-flex

  Read the article

• How to install Eclipse + PHP Development Tools (PDT) + Debugger on Mac in The Year 2010

  - by aphex5

  I had a lot of trouble installing Eclipse and PDT on my system. It took two days, largely because all the tutorials I could find were out of date (written in 2008, it's 2010 now) and various steps they included were no longer necessary, broken, or irrelevant. I wanted to write my process here so it could be improved upon (via wiki) as time goes on. Install Eclipse without PHP plugin ("Eclipse Classic"). This will give you a complete Eclipse, which I find preferable, as the UI is more fleshed out (e.g. you get a default list of Perspectives, which helps you understand what Perspectives are.) Install PDT SDK with the Help Install New Software menu item. You'd think you'd be done here, but if you try to run something, it'll fail complaining of not having a debugger. Install the Zend Debugger. It'll fail if you try to use the Install New Software option, as many tutorials suggest ("No repository found containing osgi.bundle.org.zend.php.debug.debugger.5.3.7.v20091116".) Instead, download it from http://www.zend.com/en/community/pdt, and manually copy the features/ and plugins/ directory into your Eclipse install (these instructions are not written anywhere). Restart Eclipse Monkey with preferences for a while -- if you followed a previous tutorial and tried to manually add your php executable to Eclipse prefs (/usr/bin/php), remove it (PHP PHP Executables). Set one of the Zend Debugger executables to the default. If you've already tried to execute a .php file, remove the existing "Run" profile you (maybe weren't aware that you) created (Run Debug Configurations...). Eclipse works! You should be able to run a .php file as a script just fine.

  Read the article

• distributed, fault-tolerant network block device

  - by gucki

  I'm looking for a distributed, fault-tolerant network storage system which exposes block devices (not filesystems) on the clients. A client's block device should write simultaneously to several storage nodes A client's block device should not fail as long as not all storage nodes backing it went down The master should automatically redistribute storages' data when a storage node fails or gets added/ removed A single master (which is for metadata only) is fine So ideally the architecture would be very similar to moosefs (http://www.moosefs.org/) but instead of exposing a real filesystem mounted using a fuse client it'd expose block devices on the clients. I know of iscsi and drbd but both don't seem to offer what I'm looking for. Or am I missing something?

  Read the article

< Previous Page | 470 471 472 473 474 475 476 477 478 479 480 481  | Next Page >