{"id":794,"date":"2011-11-25T10:34:24","date_gmt":"2011-11-25T18:34:24","guid":{"rendered":"https:\/\/blogs.plm.automation.siemens.com\/t5\/Polarion-Blog\/Subversion-1-7-Zippy-but-should-you-upgrade\/ba-p\/380690"},"modified":"2026-03-26T05:30:54","modified_gmt":"2026-03-26T09:30:54","slug":"subversion-1-7-zippy-but-should-you-upgrade","status":"publish","type":"post","link":"https:\/\/blogs.sw.siemens.com\/polarion\/subversion-1-7-zippy-but-should-you-upgrade\/","title":{"rendered":"Subversion 1.7: Zippy, but should you upgrade?"},"content":{"rendered":"<p><EM>by Lutz Dornbusch, Senior Consultant, Polarion Professional Services<\/EM><\/p>\n<p><STRONG><IMG class=\"alignright size-full wp-image-1773\" style=\"padding-left: 10px;\" title=\"Subversion 1.7 - Zippy but should you upgrade?\" src=\"http:\/\/community.plm.automation.siemens.com\/legacyfs\/online\/siemensplm_blogs\/2011\/11\/svn17-zippy.jpg\" alt=\"Racecar with Subversion 1.7 paint job\" width=\"300\" height=\"200\" \/>Subversion 1.7<\/STRONG> is a release mostly devoid of new features, but with many <STRONG>performance improvements<\/STRONG>. From some users\u2019 highly appreciated WC-NG to mostly invisible HTTPv2, the question is what to expect from these improvements.<\/p>\n<p>Because Polarion\u2019s application lifecycle management products are built on top of Subversion, making SVN a core issue for our customers, we felt the need to highlight and benchmark some of the most visible performance boosts of SVN 1.7. I conducted tests with several key SVN operations and found some results impressive enough to suggest that upgrading to SVN is probably worthwhile for most users. But there are some important caveats as well, especially on the client side (at the time of this writing).<\/p>\n<p>In this article, I&#8217;ll share the <STRONG>results of my performance evaluation<\/STRONG> of Subversion 1.7 and <STRONG>recommendations for upgrading<\/STRONG>. If you like, you can <STRONG><A href=\"http:\/\/www.polarion.com\/downloads\/whitepaper_svn17.pdf\" rel=\"nofollow noopener noreferrer\">download a PDF of this article<\/A><\/STRONG> (open download, no registration).<\/p>\n<p>Before you start reading &#8211; some handy links I&#8217;d like to share with you:<br \/>\n<UL><br \/>\n\t<LI><A href=\"http:\/\/www.polarion.com\/products\/requirements\/index.php\" rel=\"nofollow noopener noreferrer\">Requirements Management with Polarion REQUIREMENTS<\/A><\/LI><br \/>\n\t<LI><A href=\"http:\/\/www.polarion.com\/products\/alm\/index.php\" rel=\"nofollow noopener noreferrer\">Application Lifecycle Management with Polarion ALM<\/A><\/LI><br \/>\n\t<LI><A href=\"http:\/\/www.polarion.com\/products\/automotive\/functional-safety-iso-26262-automotive-spice-cmmi.php\" rel=\"nofollow noopener noreferrer\">Polarion Solutions for Automotive OEMs and Suppliers<\/A><\/LI><br \/>\n\t<LI><A href=\"http:\/\/www.polarion.com\/products\/medical\/index.php\" rel=\"nofollow noopener noreferrer\">Polarion Solutions for Medical Device Manufacturing<\/A><\/LI><br \/>\n\t<LI><A href=\"http:\/\/www.polarion.com\/products\/svn\/index.php\" rel=\"nofollow noopener noreferrer\">Polarion Tools and Services for Subversion<\/A><\/LI><br \/>\n\t<LI><A href=\"http:\/\/www.polarion.com\/company\/contact\/contact.php\" rel=\"nofollow noopener noreferrer\">Contact Polarion Software<\/A><\/LI><br \/>\n<\/UL><br \/>\n<H2>Measurement Procedure<\/H2><br \/>\nI measured mostly on a small 1.6 GHz dual-core laptop. The reason is that even if I have faster hardware for production, I want to see performance under low-tech conditions. But I also included some datasets on a high-end computer to compare improvements. I also tested \u201cWindows only\u201d, as this is the OS our customers are using most.<\/p>\n<p>I usually performed all commands 10 times, threw out the best and worst performance times, and calculated the average of the rest. On some tests, like <CODE>svn log<\/CODE> when caching jumps in, I used only the first (much higher) value. I used a locally-installed Apache to circumvent nearly all network latency. Since that is different everywhere and always, I decided to push it out of the equation. I also performed some checkouts on a SSD to rule out HD performance effects.<br \/>\n<H3>The Candidate Commands<\/H3><br \/>\nI was only interested in these three SVN commands:<br \/>\n<UL><br \/>\n\t<LI style=\"font-weight: bold;\"><CODE>svn checkout<\/CODE><\/LI><br \/>\n\t<LI style=\"font-weight: bold;\"><CODE>svn log<\/CODE><\/LI><br \/>\n\t<LI style=\"font-weight: bold;\"><CODE>svn status<\/CODE><\/LI><br \/>\n<\/UL><br \/>\nThese three commands are the ones Subversion users most often blame for their low performance.<\/p>\n<p>Note that <CODE>svn status<\/CODE> operates without a server connection so it is HD-bound and therefore caching is significantly affecting the measurement, which is why I used the first measured value.<br \/>\n<H3>Test Machine Specs<\/H3><br \/>\n<UL><br \/>\n\t<LI>CPU: 1.6 GHz Dual core Centrino<\/LI><br \/>\n\t<LI>RAM: 2.5 GB<\/LI><br \/>\n\t<LI>HD: 2.5\u201d 120 GB\/7200rpm<\/LI><br \/>\n\t<LI>Apache httpd: 2.2.21 (std. Windows build) on localhost<\/LI><br \/>\n\t<LI>SVN 1.6: Subversion 1.6.13<\/LI><br \/>\n\t<LI>SVN 1.7: Subversion 1.7.0<\/LI><br \/>\n<\/UL><br \/>\n<H3>The Test Repository<\/H3><br \/>\nI used a Polarion repository with a lot of projects and customer-specific configurations for my performance checks. This exemplifies Subversion\u2019s worst case scenario: a huge number of small files in another huge number of folders. As you see I have nearly as many folders as files and the average file size is about 3.2 KB.<br \/>\n<UL><br \/>\n\t<LI>Repository Size: 32.454.371 Bytes (30.9 MB)<\/LI><br \/>\n\t<LI>Revisions : 2325 revisions<\/LI><br \/>\n\t<LI>Number of Files in HEAD: 5509 files<\/LI><br \/>\n\t<LI>Number of Folders in HEAD: 4267 folders<\/LI><br \/>\n\t<LI>Overall File Size in HEAD: 18.2MB<\/LI><br \/>\n<\/UL><br \/>\n<H2>Test #1: <CODE>svn checkout<\/CODE><\/H2><br \/>\nHere are the results of a first test run in the environment I have already described:<\/p>\n<p><A href=\"http:\/\/community.plm.automation.siemens.com\/legacyfs\/online\/siemensplm_blogs\/2011\/11\/test1.png\" rel=\"nofollow noopener noreferrer\"><IMG class=\"alignnone size-medium wp-image-1764\" title=\"Results of test of svn checkout, SVN 1.6 and SVN 1.7\" src=\"http:\/\/community.plm.automation.siemens.com\/legacyfs\/online\/siemensplm_blogs\/2011\/11\/test1-300x152.png\" alt=\"Graph of test results\" width=\"300\" height=\"152\" \/><\/A><\/p>\n<p>We then made another checkout on a much faster machine with SSD (Intel core i7-2600, 8GB, Intel 320m 160Gb SSD). These are the results:<\/p>\n<p><A href=\"http:\/\/community.plm.automation.siemens.com\/legacyfs\/online\/siemensplm_blogs\/2011\/11\/test2.png\" rel=\"nofollow noopener noreferrer\"><IMG class=\"alignnone size-medium wp-image-1766\" title=\"Next results of svn checkout on different system, SVN 1.6, SVN 1.7\" src=\"http:\/\/community.plm.automation.siemens.com\/legacyfs\/online\/siemensplm_blogs\/2011\/11\/test2-300x145.png\" alt=\"Graph of another test result\" width=\"300\" height=\"145\" \/><\/A><\/p>\n<p>You can see the usage of SSD increases the performance dramatically. (Note that server and repo was sitting on a spinning 7200 rpm HD, only working copy was writing on SSD). The ratio is somewhat equal to the slower case above: SVN1.7 is much faster, even with a SVN1.6 server.<br \/>\n<H3>Analysis<\/H3><br \/>\nThese numbers are quite impressive! It seems that SVN1.7 is more than twice as fast on an initial checkout even if you are using only the SVN1.7 client. This makes an upgrade very recommendable. If you are using a SVN1.7 Server, the improvements are even higher.<\/p>\n<p>The reason for this performance impact is two-fold:<\/p>\n<p><STRONG>1.&nbsp;&nbsp;&nbsp;The new working copy format does not need to write so many files \/ folders<\/STRONG><\/p>\n<p>The new working copy format holds all SVN data on the topmost folder. No more .svn folders in each directory! As you see on the table, there is no improvement on working copy size though. However, you need many less folders(something around 260 folders max), while we had much, much more folders in old working copy format. Not so extreme, but still impressive, is the filecount: SVN1.7 will store the same files only once. In my Polarion repo are a lot files equal (mainly XML-files), so we end up with not twice of the file count. In Subversion 1.6 the file size roughly quadrupled:<\/p>\n<p><A href=\"http:\/\/community.plm.automation.siemens.com\/legacyfs\/online\/siemensplm_blogs\/2011\/11\/test3.png\" rel=\"nofollow noopener noreferrer\"><IMG class=\"alignnone size-medium wp-image-1769\" title=\"Comparison of working copy size, SVN 1.6 and SVN 1.7\" src=\"http:\/\/community.plm.automation.siemens.com\/legacyfs\/online\/siemensplm_blogs\/2011\/11\/test3-300x108.png\" alt=\"SVN working copy comparison graph, SVN 1.6 and SVN 1.7\" width=\"300\" height=\"108\" \/><\/A><\/p>\n<p><STRONG>2.&nbsp;&nbsp;&nbsp;The new HTTPv2 Protocol uses far less handshakes and so reduces Apache overhead<\/STRONG><\/p>\n<p>Here is an Apache log of a checkout of a single file (client 1.6 \/ server 1.6):<br \/>\n<DIV style=\"margin: 12px 1.5em; background-color: #eeeeee; color: #000000; font-family: monospace; padding: 8px 1em; width: 75%; font-size: 120%;\"><\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;OPTIONS \/repo\/.dms\/fields HTTP\/1.1&#8221; 200 143<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/.dms\/fields HTTP\/1.1&#8221; 207 353<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/!svn\/vcc\/default HTTP\/1.1&#8221; 207 244<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/!svn\/bln\/2325 HTTP\/1.1&#8221; 207 257<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/.dms\/fields HTTP\/1.1&#8221; 207 353<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/!svn\/vcc\/default HTTP\/1.1&#8221; 207 244<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/!svn\/bln\/2325 HTTP\/1.1&#8221; 207 257<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/.dms\/fields HTTP\/1.1&#8221; 207 353<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/!svn\/vcc\/default HTTP\/1.1&#8221; 207 257<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/!svn\/bc\/2325\/.dms\/fields HTTP\/1.1&#8221; 207 359<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;OPTIONS \/repo\/.dms\/fields HTTP\/1.1&#8221; 200 143<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/.dms\/fields HTTP\/1.1&#8221; 207 353<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/.dms\/fields HTTP\/1.1&#8221; 207 353<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/!svn\/vcc\/default HTTP\/1.1&#8221; 207 244<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/!svn\/bln\/2325 HTTP\/1.1&#8221; 207 257<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/.dms\/fields HTTP\/1.1&#8221; 207 353<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/!svn\/vcc\/default HTTP\/1.1&#8221; 207 244<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/!svn\/bln\/2325 HTTP\/1.1&#8221; 207 257<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;REPORT \/repo\/!svn\/vcc\/default HTTP\/1.1&#8221; 200 1067<\/p>\n<p><\/DIV><br \/>\nHere is the same checkout on a SVN1.7 client: (client 1.7 \/ server 1.6):<br \/>\n<DIV style=\"margin: 12px 1.5em; background-color: #eeeeee; color: #000000; font-family: monospace; padding: 8px 1em; width: 75%; font-size: 120%;\"><\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;OPTIONS \/repo\/.dms\/fields HTTP\/1.1&#8221; 200 143<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/.dms\/fields HTTP\/1.1&#8221; 207 353<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/!svn\/vcc\/default HTTP\/1.1&#8221; 207 244<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/!svn\/bln\/2325 HTTP\/1.1&#8221; 207 257<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/.dms\/fields HTTP\/1.1&#8221; 207 353<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/!svn\/vcc\/default HTTP\/1.1&#8221; 207 257<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/!svn\/bc\/2325\/.dms\/fields HTTP\/1.1&#8221; 207 359<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;OPTIONS \/repo\/.dms\/fields HTTP\/1.1&#8221; 200 143<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/.dms\/fields HTTP\/1.1&#8221; 207 353<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/.dms\/fields HTTP\/1.1&#8221; 207 353<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/!svn\/vcc\/default HTTP\/1.1&#8221; 207 244<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/!svn\/bln\/2325 HTTP\/1.1&#8221; 207 257<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;REPORT \/repo\/!svn\/vcc\/default HTTP\/1.1&#8221; 200 1067<\/p>\n<p><\/DIV><br \/>\nSo even on a 1.6 Server there is a reduced number of http transactions (19 vs. 13), but if you upgrade to a 1,7 Server you got the full power of the new HTTPv2 Protocol(client 1.7 \/ server 1.7):<br \/>\n<DIV style=\"margin: 12px 1.5em; background-color: #eeeeee; color: #000000; font-family: monospace; padding: 8px 1em; width: 75%; font-size: 120%;\"><\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;OPTIONS \/repo\/.dms\/fields HTTP\/1.1&#8221; 200 143<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;OPTIONS \/repo\/.dms\/fields HTTP\/1.1&#8221; 200 143<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/!svn\/rvr\/2325\/.dms\/fields HTTP\/1.1&#8221; 207 359<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;OPTIONS \/repo\/.dms\/fields HTTP\/1.1&#8221; 200 143<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;OPTIONS \/repo\/.dms\/fields HTTP\/1.1&#8221; 200 143<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;REPORT \/repo\/!svn\/me HTTP\/1.1&#8221; 200 1099<\/p>\n<p><\/DIV><br \/>\n<STRONG>Only 6 HTTP requests!<\/STRONG><br \/>\n<H2>Test #2: <CODE>show log all<\/CODE><\/H2><br \/>\n<IMG class=\"alignnone size-full wp-image-1778\" title=\"Results from show log, SVN 1.6, SVN 1.7\" src=\"http:\/\/community.plm.automation.siemens.com\/legacyfs\/online\/siemensplm_blogs\/2011\/11\/show_log_all.png\" alt=\"Table of test results from show log, SVN 1.6, SVN 1.7\" width=\"503\" height=\"80\" \/><br \/>\n<H3>Analysis<\/H3><br \/>\nI did not get the reason why logging is slower between a SVN 1.6 client and a SVN 1.7 server. Although this would be something interesting to investigate, I felt it is probably not one of the more common usage scenarios and so didn\u2019t take the time to delve into it. So no big news here: performance is quite the same on my machine, but it seems that the HTTP round-trips have been reduced as well.<\/p>\n<p>Log message for SVN 1.6 (client 1.6 \/ server 1.6):<br \/>\n<DIV style=\"margin: 12px 1.5em; background-color: #eeeeee; color: #000000; font-family: monospace; padding: 8px 1em; width: 75%; font-size: 120%;\"><\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;OPTIONS \/repo HTTP\/1.1&#8221; 200 143<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo HTTP\/1.1&#8221; 207 337<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/!svn\/vcc\/default HTTP\/1.1&#8221; 207 244<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/!svn\/bln\/2325 HTTP\/1.1&#8221; 207 257<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo HTTP\/1.1&#8221; 207 337<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/!svn\/vcc\/default HTTP\/1.1&#8221; 207 244<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/!svn\/bln\/2325 HTTP\/1.1&#8221; 207 257<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo HTTP\/1.1&#8221; 207 337<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/!svn\/vcc\/default HTTP\/1.1&#8221; 207 244<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/!svn\/bln\/2325 HTTP\/1.1&#8221; 207 257<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo HTTP\/1.1&#8221; 207 337<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;PROPFIND \/repo\/!svn\/vcc\/default HTTP\/1.1&#8221; 207 257<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;REPORT \/repo\/!svn\/bc\/2325 HTTP\/1.1&#8221; 200 36418<\/p>\n<p><\/DIV><br \/>\nAnd here is the SVN 1.7 log (client 1.7 \/ server 1.7):<br \/>\n<DIV style=\"margin: 12px 1.5em; background-color: #eeeeee; color: #000000; font-family: monospace; padding: 8px 1em; width: 75%; font-size: 120%;\"><\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;OPTIONS \/repo HTTP\/1.1&#8221; 200 143<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;OPTIONS \/repo HTTP\/1.1&#8221; 200 143<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;OPTIONS \/repo HTTP\/1.1&#8221; 200 143<\/p>\n<p>127.0.0.1 &#8211; &#8211;  &#8220;REPORT \/repo\/!svn\/rvr\/2325 HTTP\/1.1&#8221; 200 3641<\/p>\n<p><\/DIV><br \/>\n<H2>Test #3: <CODE>svn status<\/CODE><\/H2><br \/>\n<A href=\"http:\/\/community.plm.automation.siemens.com\/legacyfs\/online\/siemensplm_blogs\/2011\/11\/svn_status.png\" rel=\"nofollow noopener noreferrer\"><IMG class=\"alignnone size-full wp-image-1781\" title=\"Results of svn status opertaion, SVN 1.6, SVN 1.7\" src=\"http:\/\/community.plm.automation.siemens.com\/legacyfs\/online\/siemensplm_blogs\/2011\/11\/svn_status.png\" alt=\"Table of test results on svn status, SVN 1.6, SVN 1.7\" width=\"498\" height=\"79\" \/><\/A><br \/>\n<H3>Analysis<\/H3><br \/>\nOn the local SVN status command, the new working copy structure strikes again: the optimized structure in SVN 1.7 clocks in again more than twice as fast compared to SVN 1.6.<\/p>\n<p>Note that I used the whole repository as working copy, and changed around 10 to 20 files deep in the nested folders.<br \/>\n<H2>Conclusions<\/H2><br \/>\n<STRONG>Subversion 1.7 seems to dramatically improve the performance.<\/STRONG> The new working copy architecture is much faster, and the overall usage is much better: no <CODE>.svn<\/CODE> folders everywhere, so you can easily copy files from your working copy to other places.<\/p>\n<p>We can definitely recommend to upgrade at least your clients to SVN 1.7 as soon as possible!<\/p>\n<p>However, it\u2019s important to keep in mind that the <STRONG>new client\u2019s working copy format is incompatible with older (SVN1.6 ) clients<\/STRONG>. So make sure to update all clients you are using in your tool chain (e.g. TortoiseSVN, <A href=\"http:\/\/www.polarion.com\/products\/svn\/subversive.php\" rel=\"nofollow noopener noreferrer\">Subversive<\/A>, smartSVN, etc.) If some of your clients don\u2019t yet support Subversion 1.7, then you will need to wait to upgrade until they do.<\/p>\n<p><STRONG><EM>EDITOR&#8217;S NOTE:<\/EM><br \/>\n<\/STRONG>??<EM>Polarion  will officially support SVN 1.7 server in Polarion 2012, scheduled for release in Q1-2012. However, you can already upgrade your SVN clients like TortoiseSVN  to benefit from improved SVN performance.<\/EM><\/p>\n<p><HR \/><STRONG>ABOUT THE AUTHOR<\/STRONG><\/p>\n<p>Lutz Dornbusch is a senior consultant with <A href=\"http:\/\/www.polarion.com\/services\/\" rel=\"nofollow noopener noreferrer\">Polarion Software\u2019s Professional Services<\/A> group, which delivers training and consulting services for Subversion as well as the company\u2019s own products. He has more than 20 years\u2019 experience in configuration management and programming, and in his role as Subversion consultant he has successfully migrated various version control systems to Subversion. Lutz also leads the <A href=\"http:\/\/www.polarion.com\/products\/svn\/subtrain.php\" rel=\"nofollow noopener noreferrer\">SubTrain  Project<\/A> which delivers open source Subversion training materials.<\/p>\n<p><HR \/><\/p>\n","protected":false},"excerpt":{"rendered":"<p>by Lutz Dornbusch, Senior Consultant, Polarion Professional Services<\/p>\n<p>  Subversion 1.7 is a release mostly devoid of new features, but with many performance improvements. From some users\u2019 highly appr&#8230;<\/p>\n","protected":false},"author":68981,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"spanish_translation":"","french_translation":"","german_translation":"","italian_translation":"","polish_translation":"","japanese_translation":"","chinese_translation":"","footnotes":""},"categories":[1],"tags":[],"industry":[],"product":[],"coauthors":[],"class_list":["post-794","post","type-post","status-publish","format-standard","hentry","category-news"],"_links":{"self":[{"href":"https:\/\/blogs.sw.siemens.com\/polarion\/wp-json\/wp\/v2\/posts\/794","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/blogs.sw.siemens.com\/polarion\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/blogs.sw.siemens.com\/polarion\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/blogs.sw.siemens.com\/polarion\/wp-json\/wp\/v2\/users\/68981"}],"replies":[{"embeddable":true,"href":"https:\/\/blogs.sw.siemens.com\/polarion\/wp-json\/wp\/v2\/comments?post=794"}],"version-history":[{"count":1,"href":"https:\/\/blogs.sw.siemens.com\/polarion\/wp-json\/wp\/v2\/posts\/794\/revisions"}],"predecessor-version":[{"id":795,"href":"https:\/\/blogs.sw.siemens.com\/polarion\/wp-json\/wp\/v2\/posts\/794\/revisions\/795"}],"wp:attachment":[{"href":"https:\/\/blogs.sw.siemens.com\/polarion\/wp-json\/wp\/v2\/media?parent=794"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/blogs.sw.siemens.com\/polarion\/wp-json\/wp\/v2\/categories?post=794"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/blogs.sw.siemens.com\/polarion\/wp-json\/wp\/v2\/tags?post=794"},{"taxonomy":"industry","embeddable":true,"href":"https:\/\/blogs.sw.siemens.com\/polarion\/wp-json\/wp\/v2\/industry?post=794"},{"taxonomy":"product","embeddable":true,"href":"https:\/\/blogs.sw.siemens.com\/polarion\/wp-json\/wp\/v2\/product?post=794"},{"taxonomy":"author","embeddable":true,"href":"https:\/\/blogs.sw.siemens.com\/polarion\/wp-json\/wp\/v2\/coauthors?post=794"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}