Commit message (Collapse) | Author | Age | Files | Lines | ||
---|---|---|---|---|---|---|
... | ||||||
* | | Add feed block status to ClusterStateBundle in C++. | Geir Storli | 2021-01-18 | 3 | -6/+111 | |
|/ | ||||||
* | Merge pull request #16014 from vespa-engine/balder/disable-valgrind | Henning Baldersheim | 2021-01-13 | 1 | -5/+1 | |
|\ | | | | | Reduce cost as test from 400 to below 100 | |||||
| * | Update CMakeLists.txt | Henning Baldersheim | 2021-01-13 | 1 | -1/+1 | |
| | | ||||||
| * | No need to set cost explicit. | Henning Baldersheim | 2021-01-12 | 1 | -1/+1 | |
| | | ||||||
| * | Disable valgrind as this test takes 360s with and 3s without. | Henning Baldersheim | 2021-01-12 | 1 | -5/+1 | |
| | | ||||||
* | | - Factor out insert() to a separate method. | Henning Baldersheim | 2021-01-12 | 10 | -122/+103 | |
| | | | | | | | | | | | | | | | | | | - Inline Node::Node. - Remove virtuality of Node - Inline State::oneOf - Microoptimize ClusterState::getNodeState. This brings the runtime down from 2.23s to 1.68s And with valgrind from 130s to 85s. In total with the other commits in this PR it is now down from 2.95s to 1.68, and from 360s to 85s with valgrind. | |||||
* | | Allocate once and reuse. | Henning Baldersheim | 2021-01-12 | 1 | -21/+22 | |
| | | ||||||
* | | Replace the use of std::list with std::vector. This avoids a lof of new/delete. | Henning Baldersheim | 2021-01-12 | 1 | -28/+16 | |
|/ | | | | | The number of elements that are moved around is small and significantly less than the gain by more efficient memory management. | |||||
* | Eliminate inlining warnings. | Tor Egge | 2020-12-04 | 2 | -0/+4 | |
| | ||||||
* | Add noexcept | Henning Baldersheim | 2020-11-26 | 2 | -11/+11 | |
| | ||||||
* | As we have have now removed the expensive Route member we can further ↵ | Henning Baldersheim | 2020-11-26 | 2 | -20/+36 | |
| | | | | | | | | | | compact the message objects. - Compact StorageMessageAddress to 16 bytes by - using reference to cluster name. - Use small enums for protocol and node type. - Avoid having StorageMessage as separate allocation. - Avoid default values | |||||
* | Remove disk distribution from C++ part of vdslib. | Tor Egge | 2020-11-05 | 3 | -48/+4 | |
| | ||||||
* | Remove diskstate from vdslib. | Tor Egge | 2020-11-03 | 12 | -807/+20 | |
| | ||||||
* | Use std::mutex and std:.condition_variable and GC some unused code. | Henning Baldersheim | 2020-10-13 | 8 | -571/+0 | |
| | ||||||
* | Use vespalib::Lock -> std::mutex | Henning Baldersheim | 2020-10-08 | 1 | -6/+5 | |
| | ||||||
* | - GC unused code. | Henning Baldersheim | 2020-10-07 | 7 | -339/+0 | |
| | | | | - vespalib::Lock -> std::mutex | |||||
* | Remove broken copy and move constructor and assignment operatoos on ↵ | Henning Baldersheim | 2020-10-07 | 3 | -11/+4 | |
| | | | | | | vespalib::Lock and vespalib::Monitor. Also repair broken usages of the same. | |||||
* | Use noexcept as advise by gcc -Wnoexcept | Henning Baldersheim | 2020-10-06 | 2 | -2/+2 | |
| | ||||||
* | Annotate some expensive tests to allow them to start first when running in ↵ | Arnstein Ressem | 2020-09-02 | 1 | -0/+1 | |
| | | | | parallel. | |||||
* | Check _node2Group array size in Distribution. | Tor Egge | 2020-08-20 | 1 | -1/+1 | |
| | ||||||
* | Stick to junit for simple test. | Henning Baldersheim | 2020-08-11 | 1 | -16/+19 | |
| | ||||||
* | Consolidate search for GTest. | Tor Egge | 2020-07-07 | 6 | -7/+0 | |
| | ||||||
* | Improvements to handling of cluster removal | Jon Bratseth | 2020-07-02 | 1 | -5/+2 | |
| | | | | | | | | | - Don't change health status to "initializing" when creating a new VipStatus, as 'initializing' now requires all clusters to be up to transition to 'up', which means that if we're already up but are missing a cluster we'll go from 'up' to 'initializing' and stay there. - Forget up/down status for removed clusters. - Nicer logging on ignorable reconfiguration errors. | |||||
* | Merge pull request #13742 from ↵ | Geir Storli | 2020-06-30 | 6 | -6/+13 | |
|\ | | | | | | | | | vespa-engine/toregge/use-find-package-to-find-gtest Use find_package to find gtest library. | |||||
| * | Use find_package to find gtest library. | Tor Egge | 2020-06-29 | 6 | -6/+13 | |
| | | ||||||
* | | Only sort once during ideal group calculations | Tor Brede Vekterli | 2020-06-29 | 2 | -24/+127 | |
|/ | | | | | | | | | Avoids invoking `std::sort` O(n) times in favor of just once. Benchmark for 150 groups of 1 node each: * Before: 0.0004381478 seconds per invocation * After: 0.0000377917 seconds per invocation | |||||
* | Precompute node 2 group mapping config time. | Henning Baldersheim | 2020-06-26 | 3 | -18/+25 | |
| | ||||||
* | Nested namespace | Henning Baldersheim | 2020-06-26 | 2 | -13/+7 | |
| | ||||||
* | 0 -> nullptr | Henning Baldersheim | 2020-06-26 | 1 | -1/+1 | |
| | ||||||
* | Use std::move | Henning Baldersheim | 2020-06-26 | 2 | -5/+6 | |
| | ||||||
* | Modernize to c++11 style loops. | Henning Baldersheim | 2020-06-26 | 3 | -35/+22 | |
| | ||||||
* | Revert "Only sort once during ideal group calculations" | Arnstein Ressem | 2020-06-26 | 2 | -75/+24 | |
| | ||||||
* | Only sort once during ideal group calculations | Tor Brede Vekterli | 2020-06-25 | 2 | -24/+75 | |
| | | | | | | | | | Avoids invoking `std::sort` O(n) times in favor of just once. Benchmark for 150 groups of 1 node each: * Before: 0.0004381478 seconds per invocation * After: 0.0000377917 seconds per invocation | |||||
* | Don't write outside container in vdslib distribution test. | Tor Egge | 2020-04-23 | 1 | -19/+18 | |
| | ||||||
* | Add noexcept to move constructors and operators | Henning Baldersheim | 2020-01-21 | 2 | -4/+4 | |
| | ||||||
* | GC a load of unused code. ByteBuffer towards read only. | Henning Baldersheim | 2020-01-20 | 7 | -68/+69 | |
| | ||||||
* | FastOS_THread::Sleep -> std::chrono::sleep_for | Henning Baldersheim | 2019-12-04 | 1 | -2/+4 | |
| | | | | | Renamed Timer -> ScheduledExecutor. Do not include thread.h when not needed in header files. | |||||
* | Remove unused vespa/vespalib/util/regexp.h include in vdslib. | Tor Egge | 2019-10-07 | 1 | -1/+0 | |
| | ||||||
* | Use the get() api correctly. The value is returned. | Henning Baldersheim | 2019-10-01 | 1 | -21/+2 | |
| | ||||||
* | Remove v6 and v7 from java side. | Henning Baldersheim | 2019-08-21 | 1 | -21/+0 | |
| | ||||||
* | Remove visitor ordering and order selection. | Henning Baldersheim | 2019-08-09 | 4 | -93/+0 | |
| | ||||||
* | Remove the use and testing of legacy groupdoc/userdoc/orderdoc document ids. | Henning Baldersheim | 2019-08-09 | 1 | -40/+0 | |
| | ||||||
* | Use int64_t for vdslib::Parameters overload. | Tor Egge | 2019-08-07 | 1 | -4/+4 | |
| | ||||||
* | Add missing space | Henning Baldersheim | 2019-08-03 | 1 | -1/+1 | |
| | ||||||
* | Update test. | Henning Baldersheim | 2019-08-03 | 1 | -1/+1 | |
| | ||||||
* | Revert "Revert "Add maxRedundancy to exception message."" | Henning Baldersheim | 2019-08-03 | 1 | -1/+3 | |
| | ||||||
* | Revert "Add maxRedundancy to exception message." | Henning Baldersheim | 2019-08-03 | 1 | -3/+1 | |
| | ||||||
* | Add maxRedundancy to exception message. | Henning Baldersheim | 2019-08-03 | 1 | -1/+3 | |
| | ||||||
* | Add missing includes. | Tor Egge | 2019-06-11 | 1 | -0/+1 | |
| | ||||||
* | Add new DB merging API to distributor BucketDatabase | Tor Brede Vekterli | 2019-06-03 | 2 | -0/+11 | |
| | | | | | | | | | | | | | | | | | | | | | | | Abstracts away how an ordered merge may be performed with the database and an arbitrary sorted bucket sequence, with any number of buckets skipped, updated or inserted as part of the merge. Such an API is required to allow efficient bulk updates of a B-tree backed database, as it is suboptimal to require constant tree mutations. Other changes: - Removed legacy mutable iteration API. Not needed with new merge API. - Const-iteration of bucket database now uses an explicit const reference entry type to avoid needing to construct a temporary entry when we can instead just point directly into the backing ArrayStore. - Micro-optimizations of node remover pass to avoid going via cluster state's node state std::map for each bucket replica entry. Now uses a precomputed bit vector. Also avoid BucketId bit reversing operations as much as possible by using raw bucket keys in more places. - Changed wording and contents of log message that triggers when buckets are removed from the DB due to no remaining nodes containing replicas for the bucket. Now more obvious what the message actually means. - Added several benchmark tests (disabled by default) |