Search Results

Search found 57986 results on 2320 pages for 'breadth first search'.

Page 350/2320 | < Previous Page | 346 347 348 349 350 351 352 353 354 355 356 357  | Next Page >

  • ldapsearch against Active Directory fails

    - by Guacamole
    I am using ldapsearch from OpenLDAP tools to search our corporate Active Directory for my email and phone number. This query is a test to ensure that I can authenticate against the domain so I can set up a linux wiki with NTLM authentication. My theory is that if I can successfully query the AD for information, then I am a step closer to getting my wiki to authenticate against AD (I have instructions to set up moin wiki under ActiveDirectory). The problem is that I can't seem to get the ldapsearch query right. I have seen many tutorials on the net that indicate that -D should be something like -D "Americas\John_Marsharll"; however, I keep getting ldap_bind: Invalid credentials (49) error messages when I use Americas\John_Marshall. The only time I get sensical results is when I query with the parameters below. However, even then, I can't figure out how to get email and phone number. [John_Marsharll@WN7-BG3YSM1 ~]$ ldapsearch -x -h 10.1.1.1 \ -b "cn=Users,dc=Americas" mail telephonenumber -D "cn=John_Marshall,dc=Americas" # extended LDIF # # LDAPv3 # base <cn=Users,dc=Americas> with scope subtree # filter: (objectclass=*) # requesting: mail telephonenumber -D cn=John_Marshall,dc=Americas # # search result search: 2 result: 32 No such object # numResponses: 1 [John_Marshall@WN7-BG3YSM1 ~]$ Can someone give me pointers on what I'm doing wrong with the ldapsearch query above? Our AD ldap server is 10.1.1.1 and the AD domain is "Americas".

    Read the article

  • grub2 error : out of disk

    - by Nostradamnit
    Hi everyone, I recently install ArchBang on a machine with Ubuntu and XP. I ran update-grub from Ubuntu and it found the new install and created an entry. However, when I try to boot it, I get: error: out of disk error: you need to load kernel first I've tried several things, including adding a new entry in 40_custom, but nothing changes. Here are the entries I have: default found by update-grub ### BEGIN /etc/grub.d/30_os-prober ### menuentry "ArchBang Linux (on /dev/sda4)" { insmod part_msdos insmod ext2 set root='(hd0,msdos4)' search --no-floppy --fs-uuid --set 75f96b44-3a8f-4727-9959-d669b9244f2a linux /boot/vmlinuz26 root=/dev/sda4 rootfstype=ext4 ro xorg=vesa quiet nomodeset swapon initrd /boot/kernel26.img } menuentry "ArchBang Linux Fallback (on /dev/sda4)" { insmod part_msdos insmod ext2 set root='(hd0,msdos4)' search --no-floppy --fs-uuid --set 75f96b44-3a8f-4727-9959-d669b9244f2a linux /boot/vmlinuz26 root=/dev/sda4 rootfstype=ext4 ro xorg=vesa quiet nomodeset swapon initrd /boot/kernel26-fallback.img } ### END /etc/grub.d/30_os-prober ### custom entry in 40_custom based on various ideas found on the internets menuentry "ArchBang Linux (on /dev/sda4)" { insmod part_msdos insmod ext2 set root='(hd0,msdos4)' search --no-floppy --fs-uuid --set 75f96b44-3a8f-4727-9959-d669b9244f2a linux /boot/vmlinuz26 root=/dev/disk/by-uuid/75f96b44-3a8f-4727-9959-d669b9244f2a rootfstype=ext4 ro xorg=vesa quiet nomodeset swapon initrd /boot/kernel26.img } I think the problem has something to do with the sda4 not being mounted at boot-time... Thanks in advance for you help, Sam

    Read the article

  • Laptop power supplies, does current matter?

    - by CodeSlave
    I have two laptops (same manufacture), with the same type of power connector. However, the power supplies/transformers are slightly different. The output on the first laptop's power supply is 15.6 V at 8.0 A. The output on the second laptop's power supply is 15.6 V at 5A. Clearly the voltages are the same, but the currents are different. I assume the second laptop's power supply can not be used on the first, because it can't supply enough power to the laptop. However, can the first laptop's power supply be safely used on the second laptop?

    Read the article

  • Is there a reason the partition tool GParted doesn't show a percentage finish number initially?

    - by Jian Lin
    I am using GParted more, and it seems to do a reliable work. I just wonder why if the tasks is to resize a 250GB partition to 190GB, and then create a new partition, the first 10, 15 minutes, there is only a blue bar moving left and right, but there won't be an indicator showing how many percent is done. Then after that 10 to 15 minutes, it does show 1:05:00 left to finish the job. Update: at first I thought there is no percentage or time remaining at all... but after waiting for 10, 15 mintues, it does show. I just wonder why it didn't show at first.

    Read the article

  • Multi-Document TOC showing in wrong order

    - by Jeremy DeStefano
    I had a large document that was having formatting issues, so I split it into 2 files. Chapters 1-7 are in the main doc with the TOC and a second doc has chapters 8-12. I have the following: {TOC \O "1-3" \H \Z \U} {RD \f "MCDPS Training Manual Part2.docx"} The TOC is created and has entries from both documents, however its showing the entries from Chapter 8-11 first and then Chapter 1-7. I've read that it should list them based on page numbers, but its not. Chapter 8 starts at page 121, yet its listing it first. How can I get it to show the TOC from the main doc first and then the RD?

    Read the article

  • Using Windows Explorer, how to find file names starting with a dot (period), in 7 or Vista?

    - by Chris W. Rea
    I've got a MacBook laptop in the house, and when Mac OS X copies files over the network, it often brings along hidden "dot-files" with it. For instance, if I copy "SomeUtility.zip", there will also be copied a hidden ".SomeUtility.zip" file. I consider these OS X dot-files as useless turds of data as far as the rest of my network is concerned, and don't want to leave them on my Windows file server. Let's assume these dot-files will continue to happen. i.e. Think of the issue of getting OS X to stop creating those files, in the first place, to be another question altogether. Rather: How can I use Windows Explorer to find files that begin with a dot / period? I'd like to periodically search my file server and blow them away. I tried searching for files matching ".*" but that yielded – and not unexpectedly – all files and folders. Is there a way to enter more specific search criteria when searching in Windows Explorer? I'm referring to the search box that appears in the upper-right corner of an Explorer window. Please tell me there is a way to escape my query to do what I want? (Failing that, I know I can map a drive letter and drop into a cygwin prompt and use the UNIX 'find' command, but I'd prefer a shiny easy way.)

    Read the article

  • Flow of packets in network

    - by user58859
    I can't visualize in my mind the network traffic flow. eg. If there are 15 pc's in a LAN When packet goes from router to local LAN, do it passes all the computers? Does it go to the ethernet card of every computer and those computers accept the packet based on their physical address? To which pc the packet will go first? To the nearest to the router? What happens if that first pc captures that packet(though it is not for it)? What happens when a pc broadcast a message? Do it have to generate 14 packets for all the pc's or only one packet reach to all pc's? If it is one packet and captured by first pc, how other pc's can get that? I can't imagine how this traffic is exactly flows? May be my analogy is completely wrong. Can anybody explain me this?

    Read the article

  • Word: MAC 2011, TOC on too many pages

    - by Mark
    I have a Word: MAC 2011 document where the bottom of the first 40 pages or so say "TOC: Page x". This notation appears to be in the Footer, as it is gray until I click on it (then the rest of the text goes gray instead). There is no TOC that I can see in the document, so I'm presuming someone tried to create one and messed things up. After the first 40 pages or so, all the other bottom of the page notations appear to be correct. (i.e. Chapter One, Chapter Two, etc.) How can I get those first 40 pages to be part of Chapter One rather than TOC?

    Read the article

  • Why does Windows Explorer highlight only second entry?

    - by normanius
    Why does Windows Explorer highlight the second but never the first entry if I use the keyboard for navigation? Example: let's look at a folder that contains the following entries a1 a2 a3 b1 b2 If I hit 'a' on the keyboard, the explorer highlights entry 'a2' instead of 'a1'. It works fine for 'b' with 'b1' (because it's not the first entry). Similarly, if I open a folder and use the arrow-down key to navigate then the first entry is skipped again. Why?! It's probable that I'm too stupid for this but this "feature" really annoys me!

    Read the article

  • Boot from Second SATA Drive

    - by Chris
    I have a Dell Precision 490 Workstation, and I just had my other question answered, Install Ubuntu to drive B without impacting drive A, and now I'm having a boot sequence issue. The external drive is great, boots up fine on my laptop, but how do I tell my desktop to boot from my second SATA drive and not the first SATAdrive. My drive configuration as follows SATA-0: Windows SATA-1: DVDR SATA-2: Ubuntu When I choose the boot menu, the option I have is "Internal Hard Drive". I assume it searches all drives, and loads the first bootable one it finds (which happens to be Windows), but I'd like to be able to select the drive from a list. Has anyone experienced this? Is possible without disabling the first hard drive in the BIOS?

    Read the article

  • Ubuntu IP Configuration - multiple subnets & interfaces

    - by HaydnWVN
    Have a 'new' mailserver running postfix on Ubuntu. We are having some problems configuring the subnets & interfaces. Basically 2 subnets (.253. & .254.) need to be connected through the 3rd subnet (.252.) where the Router is residing. # This file describes the network interfaces available on your system # and how to activate them. For more information, see interfaces(5). # The loopback network interface auto lo iface lo inet loopback # The primary network interface auto eth0 iface eth0 inet static address 10.62.254.199 netmask 255.255.0.0 network 10.62.254.0 broadcast 10.62.255.255 #gateway 10.62.252.138 # dns-* options are implemented by the resolvconf package, if installed dns-nameservers 10.62.252.138 dns-search ***.com auto eth1 iface eth1 inet static address 10.62.253.199 netmask 255.255.0.0 network 10.62.253.0 broadcast 10.62.255.255 #gateway 10.62.252.138 #dns-nameservers 10.62.254.199 10.62.253.199 10.62.252.199 dns-nameservers 10.62.252.138 dns-search ***.com auto eth2 iface eth2 inet static address 10.62.252.199 netmask 255.255.0.0 network 10.62.252.0 broadcast 10.62.255.255 gateway 10.62.252.138 #dns-nameservers 10.62.254.199 10.62.253.199 10.62.252.199 dns-search ***.com I have an external support company who are looking into this (they built and configured this server), but it's taking far too long... So I'm looking to highlight the mistake!

    Read the article

  • Pre-load MS Windows right-click menus and Start menu at startup

    - by Steve
    Hello brainy people. On my WinXP SP3 laptop (1.4Ghz 1.2GB ram), after I first log in, when I right-click in Windows Explorer and choose New, the submenu can take up to 15 seconds to load, which is a pain in the ass when you want to do a quick easy operation. After the submenu has loaded the first time, subsequent loads perform instantly, obviously as the menu has been cached. My question is: can these right-click menus (and the Start menu, which also takes some time to load the first time) be pre-loaded at Windows startup? Thanks.

    Read the article

  • Thunderbird: export email account settings

    - by zpea
    I'd like to create a new profile for Thunderbird using the same mail accounts I already configured in my old profile. As it is quite a number of accounts, it would be great to have a way to export/import them instead of writing down the settings just to fill in again in the new profile. Using web search and search here I mainly found following suggestions that do not match what I need: Copy the whole profile: Not possible for me as I don't want to copy other settings, the downloaded mail data etc. and the old profile broke when running out of space in the home folder anyway. Use mozBackup: There seem to be several programs by that name (forks?). In any case, it's Windows-only and hence no option (I am mostly on Linux and prefer platform-independent solutions anyway) Use accountex: Seems to do what I want, but it is not compatible with current Thunderbird version (supports only up to version 3.1) Posts with various tips from 4 years ago: Top results in the web search with the G. But they do not work in current versions of Thunderbird either. Did I overlook anything? After all, it doesn't sound like I was looking for something nobody ever looked for.

    Read the article

  • How Do I Enable CUPS Browsing Across A Network?

    - by David Mackintosh
    I have a CUPS server with two print queues defined. Once this was defined, all the CUPS clients on the same subnet could see the two print queues automatically, no problem. Now I have a collection of machines on a separate subnet, reachable from the first subnet by a router. How do I enable CUPS browsing on the second set of machines so that they can see the print queues defined on the first machine? Let's call the server A.B.C.7. The first subnet is A.B.C.0/24. The second subnet is A.B.D.0/24, and there is a router with arms on both networks.

    Read the article

  • Excel Matching problem with logic expression

    - by abelenky
    I have a block of data that represents the steps in a process and the possible errors: ProcessStep Status FeesPaid OK FormRecvd OK RoleAssigned OK CheckedIn Not Checked In. ReadyToStart Not Ready for Start I want to find the first Status that is not "OK". I have attempted this: =Match("<>""OK""", StatusRange, 0) which is supposed to return the index of the first element in the range that is NOT-EQUAL (<) to "OK" But this doesn't work, instead returning #N/A. I expect it to return 4 (index #4, in a 1-based index, representing that CheckedIn is the first non-OK element) Any ideas how to do this?

    Read the article

  • Convert text to table

    - by Quattro
    I would like convert text into a table. Here is a link to the text http://www.tcdb.org/public/tcdb Short example: >gnl|TC-DB|A0CIB0|1.A.17.3.1 Chromosome undetermined scaffold_19, whole genome shotgun sequence OS=Paramecium tetraurelia GN=GSPATT00007662001 PE=4 SV=1 MDDQNQPILQEQPKPKQKKPLLNTKMVKKQKMQNKKEENLREILNFYTNQVDARKFLQKM KAVVDSNQQEKKYQDDFLNPNEYNEMQDIYEDYNMGDLVIVFPNPDADGVKNPPITYKEA PLTKTNFYSKIGNVSYENDIDELCVDEMEYLRNMRNVDGEHMDQDHVKEEI >gnl|TC-DB|A0CS82|9.B.82.1.5 Chromosome undetermined scaffold_26, whole genome shotgun sequence - Paramecium tetraurelia. MIIEEQIEEKMIYKAIHRVKVNYQKKIDRYILYKKSRWFFNLLLMLLYAYRIQNIGGFYI VTYIYCVYQLQLLIDYFTPLGLPPVNLEDEEEDDDQFQNDFSELPTTLSNKNELNDKEFR PLLRTTSEFKVWQKSVFSVIFAYFCTYIPIWDIPVYWPFLFCYFFVIVGMSIRKYIKHMK KYGYTILDFTKKK I wanted to have columns for example delimited with pipe | or ; |>gnl|TC-DB|A0CIB0|1.A.17.3.1| Chromosome undetermined scaffold_19, whole genome shotgun sequence OS=Paramecium tetraurelia GN=GSPATT00007662001 PE=4 SV=1| MDDQNQPILQEQPKPKQKKPLLNTKMVKKQKMQNKKEENLREILNFYTNQVDARKFLQKM KAVVDSNQQEKKYQDDFLNPNEYNEMQDIYEDYNMGDLVIVFPNPDADGVKNPPITYKEA PLTKTNFYSKIGNVSYENDIDELCVDEMEYLRNMRNVDGEHMDQDHVKEEI I am working with Windows and I don't know how to do it I just know every row starts with > I want to substitute the first whitespace in a row with a delimiter like | or ; after the first regular expression new line in a row, I want also a delimiter everything between the regular expression first new line and > should go into a new column (it's a sequence of a protein)

    Read the article

  • Read access to Active Directory property (uSNChanged)

    - by Tom Ligda
    I have an issue with read access to the uSNChanged property when doing LDAP searches. If I do an LDAP search with a user that is a member of the Domain Admins group (UserA), I can see the uSNChanged property for every user. The problem is that if I do an LDAP search with a user (UserB) that is not a member of the Domain Admins group, I can see the uSNChanged property for some users (UserGroupA) and not for some users (UserGroupB). When I look at the users in UserGroupA and compare them to the users in UserGroupB, I see a crucial difference in the "Security" tab. The users in UserGroupA have the "Include inheritable permissions from this object's parent" unchecked. The users in UserGroupB have that option checked. I also noticed that the users in UserGroupA are users that were created earlier. The users in UserGroupB are users created recently. It's difficult to quantify, but I estimate the border between creation time between the users in UserGroupA and UserGroupB is about 6 months ago. What can cause the user creation to default to having that security property checked as opposed to unchecked? A while back (maybe around 6 months ago?) I changed the domain functional level from Windows Server 2003 to Windows Server 2008 R2. Would that have had this effect? (I can't exactly downgrade the domain functional level to test it out.) Is this security property actually the cause of the issue with read access to the uSNChanged property on LDAP searches? It seems correlated, but I'm not sure about causation. What I want in the end is for all authenticated users to have read access to the uSNChanged property for all users when doing an LDAP search. I would also be OK if I could grant read access for that property to an AD group. Then I can control access by adding members to the group.

    Read the article

  • Sudo asks for password twice with LDAP authentication

    - by Gnudiff
    I have Ubuntu 8.04 LTS machine and Windows 2003 AD domain. I have succesfully set up that I can log in with domain username and password, using domain prefix, like "domain+username". Upon login to machine it all works first try, however, for some reason when I try to sudo my logged in user, it asks for the password twice every time when I try sudo. It accepts the password after 2nd time, but not the first time. Once or twice I might think I just keep entering wrong pass the first time, but this is what happens always, any ideas of what's wrong? pam.conf is empty pam.d/sudo only includes common-auth & common-account, and common-auth is: auth sufficient pam_unix.so nullok_secure auth sufficient pam_winbind.so auth requisite pam_deny.so auth required pam_permit.so

    Read the article

  • ESX Firewall Command Troubles

    - by John
    Hi, I am working on creating some firewall rules to stop some of the SSH brute-force attacks that we have seen recently on our ESX server hosts. I have tried the following rules from the CLI to first block all SSH traffic and then allow the two ranges that I am interested in: esxcfg-firewall --ipruleAdd 0.0.0.0/0,22,tcp,REJECT,"Block_SSH" esxcfg-firewall --ipruleAdd 11.130.0.0/16,22,tcp,ACCEPT,"Allow_PUBLIC_SSH" esxcfg-firewall --ipruleAdd 10.130.0.0/16,22,tcp,ACCEPT,"Allow_PRIVATE_SSH" However, these rules are not working as intended. I know that if you do not enter the block rule first, then the allow rule will not be processed. We are now having the issue where the first entered allow rule is being ignored such that the block rule works and the last entered allow rule works. I was curious if anyone had any ideas on how I could allow a few different ranges of IP's with the esxcfg-firewall --ipruleAdd command? I am at a loss and am having a hard time locating examples or further documentation about this. Thanks in advance for your help with this.

    Read the article

  • How to convert excel individual cell values to percentage change values over time

    - by cgalloway
    I have two years of excel data showing daily share prices of a particular stock. I want to change those values to show percentage change (on a daily basis) from the zero date (ie the first day of the two year period). I know that the formula for showing daily percentage change would be (second day/first day -1) and that I can click and drag on that formula to extend over the rest of the two-year time period. The formula I want would be, basically, (each day/first day-1). Is there an easy way to automate the script so I dont have to type it out 730 times?

    Read the article

  • Asterisk Register username with special character like "@"

    - by Najibul Huq
    I am using a SIP provider that has provided me with a username like: [email protected] (Note this is only the username part) And has a numerical password. My Register string looks something like this: [email protected]:[email protected] But this is not working, as asterisk is only sending the first part +112223344 before the first @. My provider is adamant about having the full form of it. This is the first time I am facing this issue that is quite unusual for me. Please help.

    Read the article

  • Merged rows in column, bottom row fixed height

    - by Styxxy
    I've been struggling some time now with a specific problem using Tables in MS Word (2010). I have a table with 2 rows and 2 columns and the last column, the rows are merged. Now it can happen that this last cell will expand, and I would like to have the last row in the first column to be of a fixed height and the first row has to expand. What happens now is that the last row expands and the first row has a "fixed" height. A picture of the behaviour at this moment: And this is how I would like it to behave: I have been looking through all properties and settings, but I don't seem to find any option. Neither can I found anything by searching online (probably not using the exact right keywords). Any help is appreciated.

    Read the article

  • Ubuntu 12.04 on Amazon EC2: /dev/xvda1 will be checked for errors at next reboot?

    - by cwd
    I'm running the lastest Ubuntu 12.04 AMI (ami-a29943cb) from Canonical on Amazon EC2 and quite often when I log in I get the message: *** /dev/xvda1 will be checked for errors at next reboot *** I have read a bunch of documentation on this and seem to understand that every so many reboots (around 37 see Mount count / Maximum mount count below) Ubuntu wants to check a disk for errors. I can see that by using dumpe2fs -h /dev/xvda1 (reference) to get information such as: Last mounted on: / Filesystem UUID: 1ad27d06-4ecf-493d-bb19-4710c3caf924 Filesystem magic number: 0xEF53 Filesystem revision #: 1 (dynamic) Filesystem features: has_journal ext_attr resize_inode dir_index filetype needs_recovery extent flex_bg sparse_super large_file huge_file uninit_bg dir_nlink extra_isize Filesystem flags: signed_directory_hash Default mount options: (none) Filesystem state: clean Errors behavior: Continue Filesystem OS type: Linux Inode count: 524288 Block count: 2097152 Reserved block count: 104857 Free blocks: 1778055 Free inodes: 482659 First block: 0 Block size: 4096 Fragment size: 4096 Reserved GDT blocks: 511 Blocks per group: 32768 Fragments per group: 32768 Inodes per group: 8192 Inode blocks per group: 512 Flex block group size: 16 Filesystem created: Tue Apr 24 03:07:48 2012 Last mount time: Thu Nov 8 03:17:58 2012 Last write time: Tue Apr 24 03:08:52 2012 Mount count: 3 Maximum mount count: 37 Last checked: Tue Apr 24 03:07:48 2012 Check interval: 15552000 (6 months) Next check after: Sun Oct 21 03:07:48 2012 Lifetime writes: 2454 MB Reserved blocks uid: 0 (user root) Reserved blocks gid: 0 (group root) First inode: 11 Inode size: 256 Required extra isize: 28 Desired extra isize: 28 Journal inode: 8 Default directory hash: half_md4 Directory Hash Seed: 0a25e04c-6169-4d68-bfa6-a1acd8e39632 Journal backup: inode blocks Journal features: journal_incompat_revoke Journal size: 128M Journal length: 32768 Journal sequence: 0x0000158b Journal start: 1 I've tried these things to get rid of the message and usually the badblocks is what does it for me: Run this command and reboot: sudo touch /forcefsck Run badblocks to check the disk: badblocks /dev/sda1 Edit /etc/fstab and change the last "0" which is the fs_passno column accordingly and then reboot: The root filesystem should be specified with a fs_passno of 1, and other filesystems should have a fs_passno of 2. I don't understand: If this is a virtual drive shouldn't it be less prone to errors? Was the image created with one of the flags set? If not what is triggering it? Why is fs_passno set to 0 on Amazon EC2 Ubuntu images? This is not the first one that is like this.

    Read the article

  • Sublinear Extra Space MergeSort

    - by hulkmeister
    I am reviewing basic algorithms from a book called Algorithms by Robert Sedgewick, and I came across a problem in MergeSort that I am, sad to say, having difficulty solving. The problem is below: Sublinear Extra Space. Develop a merge implementation that reduces that extra space requirement to max(M, N/M), based on the following idea: Divide the array into N/M blocks of size M (for simplicity in this description, assume that N is a multiple of M). Then, (i) considering the blocks as items with their first key as the sort key, sort them using selection sort; and (ii) run through the array merging the first block with the second, then the second block with the third, and so forth. The problem I have with the problem is that based on the idea Sedgewick recommends, the following set of arrays will not be sorted: {0, 10, 12}, {3, 9, 11}, {5, 8, 13}. The algorithm I use is the following: Divide the full array into subarrays of size M. Run Selection Sort on each of the subarrays. Merge each of the subarrays using the method Sedgwick recommends in (ii). (This is where I encounter the problem of where to store the results after the merge.) This leads to wanting to increase the size of the auxiliary space needed to handle at least two subarrays at a time (for merging), but based on the specifications of the problem, that is not allowed. I have also considered using the original array as space for one subarray and using the auxiliary space for the second subarray. However, I can't envision a solution that does not end up overwriting the entries of the first subarray. Any ideas on other ways this can be done? NOTE: If this is suppose to be on StackOverflow.com, please let me know how I can move it. I posted here because the question was academic.

    Read the article

  • Parallelism in .NET – Part 4, Imperative Data Parallelism: Aggregation

    - by Reed
    In the article on simple data parallelism, I described how to perform an operation on an entire collection of elements in parallel.  Often, this is not adequate, as the parallel operation is going to be performing some form of aggregation. Simple examples of this might include taking the sum of the results of processing a function on each element in the collection, or finding the minimum of the collection given some criteria.  This can be done using the techniques described in simple data parallelism, however, special care needs to be taken into account to synchronize the shared data appropriately.  The Task Parallel Library has tools to assist in this synchronization. The main issue with aggregation when parallelizing a routine is that you need to handle synchronization of data.  Since multiple threads will need to write to a shared portion of data.  Suppose, for example, that we wanted to parallelize a simple loop that looked for the minimum value within a dataset: double min = double.MaxValue; foreach(var item in collection) { double value = item.PerformComputation(); min = System.Math.Min(min, value); } .csharpcode, .csharpcode pre { font-size: small; color: black; font-family: consolas, "Courier New", courier, monospace; background-color: #ffffff; /*white-space: pre;*/ } .csharpcode pre { margin: 0em; } .csharpcode .rem { color: #008000; } .csharpcode .kwrd { color: #0000ff; } .csharpcode .str { color: #006080; } .csharpcode .op { color: #0000c0; } .csharpcode .preproc { color: #cc6633; } .csharpcode .asp { background-color: #ffff00; } .csharpcode .html { color: #800000; } .csharpcode .attr { color: #ff0000; } .csharpcode .alt { background-color: #f4f4f4; width: 100%; margin: 0em; } .csharpcode .lnum { color: #606060; } This seems like a good candidate for parallelization, but there is a problem here.  If we just wrap this into a call to Parallel.ForEach, we’ll introduce a critical race condition, and get the wrong answer.  Let’s look at what happens here: // Buggy code! Do not use! double min = double.MaxValue; Parallel.ForEach(collection, item => { double value = item.PerformComputation(); min = System.Math.Min(min, value); }); This code has a fatal flaw: min will be checked, then set, by multiple threads simultaneously.  Two threads may perform the check at the same time, and set the wrong value for min.  Say we get a value of 1 in thread 1, and a value of 2 in thread 2, and these two elements are the first two to run.  If both hit the min check line at the same time, both will determine that min should change, to 1 and 2 respectively.  If element 1 happens to set the variable first, then element 2 sets the min variable, we’ll detect a min value of 2 instead of 1.  This can lead to wrong answers. Unfortunately, fixing this, with the Parallel.ForEach call we’re using, would require adding locking.  We would need to rewrite this like: // Safe, but slow double min = double.MaxValue; // Make a "lock" object object syncObject = new object(); Parallel.ForEach(collection, item => { double value = item.PerformComputation(); lock(syncObject) min = System.Math.Min(min, value); }); This will potentially add a huge amount of overhead to our calculation.  Since we can potentially block while waiting on the lock for every single iteration, we will most likely slow this down to where it is actually quite a bit slower than our serial implementation.  The problem is the lock statement – any time you use lock(object), you’re almost assuring reduced performance in a parallel situation.  This leads to two observations I’ll make: When parallelizing a routine, try to avoid locks. That being said: Always add any and all required synchronization to avoid race conditions. These two observations tend to be opposing forces – we often need to synchronize our algorithms, but we also want to avoid the synchronization when possible.  Looking at our routine, there is no way to directly avoid this lock, since each element is potentially being run on a separate thread, and this lock is necessary in order for our routine to function correctly every time. However, this isn’t the only way to design this routine to implement this algorithm.  Realize that, although our collection may have thousands or even millions of elements, we have a limited number of Processing Elements (PE).  Processing Element is the standard term for a hardware element which can process and execute instructions.  This typically is a core in your processor, but many modern systems have multiple hardware execution threads per core.  The Task Parallel Library will not execute the work for each item in the collection as a separate work item. Instead, when Parallel.ForEach executes, it will partition the collection into larger “chunks” which get processed on different threads via the ThreadPool.  This helps reduce the threading overhead, and help the overall speed.  In general, the Parallel class will only use one thread per PE in the system. Given the fact that there are typically fewer threads than work items, we can rethink our algorithm design.  We can parallelize our algorithm more effectively by approaching it differently.  Because the basic aggregation we are doing here (Min) is communitive, we do not need to perform this in a given order.  We knew this to be true already – otherwise, we wouldn’t have been able to parallelize this routine in the first place.  With this in mind, we can treat each thread’s work independently, allowing each thread to serially process many elements with no locking, then, after all the threads are complete, “merge” together the results. This can be accomplished via a different set of overloads in the Parallel class: Parallel.ForEach<TSource,TLocal>.  The idea behind these overloads is to allow each thread to begin by initializing some local state (TLocal).  The thread will then process an entire set of items in the source collection, providing that state to the delegate which processes an individual item.  Finally, at the end, a separate delegate is run which allows you to handle merging that local state into your final results. To rewriting our routine using Parallel.ForEach<TSource,TLocal>, we need to provide three delegates instead of one.  The most basic version of this function is declared as: public static ParallelLoopResult ForEach<TSource, TLocal>( IEnumerable<TSource> source, Func<TLocal> localInit, Func<TSource, ParallelLoopState, TLocal, TLocal> body, Action<TLocal> localFinally ) The first delegate (the localInit argument) is defined as Func<TLocal>.  This delegate initializes our local state.  It should return some object we can use to track the results of a single thread’s operations. The second delegate (the body argument) is where our main processing occurs, although now, instead of being an Action<T>, we actually provide a Func<TSource, ParallelLoopState, TLocal, TLocal> delegate.  This delegate will receive three arguments: our original element from the collection (TSource), a ParallelLoopState which we can use for early termination, and the instance of our local state we created (TLocal).  It should do whatever processing you wish to occur per element, then return the value of the local state after processing is completed. The third delegate (the localFinally argument) is defined as Action<TLocal>.  This delegate is passed our local state after it’s been processed by all of the elements this thread will handle.  This is where you can merge your final results together.  This may require synchronization, but now, instead of synchronizing once per element (potentially millions of times), you’ll only have to synchronize once per thread, which is an ideal situation. Now that I’ve explained how this works, lets look at the code: // Safe, and fast! double min = double.MaxValue; // Make a "lock" object object syncObject = new object(); Parallel.ForEach( collection, // First, we provide a local state initialization delegate. () => double.MaxValue, // Next, we supply the body, which takes the original item, loop state, // and local state, and returns a new local state (item, loopState, localState) => { double value = item.PerformComputation(); return System.Math.Min(localState, value); }, // Finally, we provide an Action<TLocal>, to "merge" results together localState => { // This requires locking, but it's only once per used thread lock(syncObj) min = System.Math.Min(min, localState); } ); Although this is a bit more complicated than the previous version, it is now both thread-safe, and has minimal locking.  This same approach can be used by Parallel.For, although now, it’s Parallel.For<TLocal>.  When working with Parallel.For<TLocal>, you use the same triplet of delegates, with the same purpose and results. Also, many times, you can completely avoid locking by using a method of the Interlocked class to perform the final aggregation in an atomic operation.  The MSDN example demonstrating this same technique using Parallel.For uses the Interlocked class instead of a lock, since they are doing a sum operation on a long variable, which is possible via Interlocked.Add. By taking advantage of local state, we can use the Parallel class methods to parallelize algorithms such as aggregation, which, at first, may seem like poor candidates for parallelization.  Doing so requires careful consideration, and often requires a slight redesign of the algorithm, but the performance gains can be significant if handled in a way to avoid excessive synchronization.

    Read the article

< Previous Page | 346 347 348 349 350 351 352 353 354 355 356 357  | Next Page >