| /rocksdb-6.9/docs/_posts/ |
| D | 2015-07-15-rocksdb-2015-h2-roadmap.markdown | 16 … to be done. We need to focus on both performance and stability. The most high priority items on a…
27 3. Reduce P99 read latency, which is high mostly because of lingering tombstones
51 …. However, there are some use-cases at Facebook that don't need expensive high-end storage. In the…
69 1. Iterate P99 can be high due to the presence of tombstones.
72 2. Write stalls can happen during high write loads.
|
| D | 2017-05-14-core-local-stats.markdown | 12 For concurrent/high-throughput workloads, cache line bouncing of atomics caused high CPU utilizatio…
36 The high CPU overhead for updating tickers and histograms corresponds well to the high cache misses.
75 … performance issues when connection count grows very high. For debugging such issues, they want hi…
|
| D | 2014-09-12-cuckoo.markdown | 12 …int lookups. The new format was built for applications which require very high point lookup rates …
46 In order to achieve high lookup speeds, we did multiple optimizations, including a cache friendly c…
|
| D | 2015-07-22-rocksdb-is-now-available-in-windows-platform.markdown | 10 …DB is successfully used by the community and various companies to achieve high throughput and volu…
|
| D | 2014-04-07-rocksdb-2-8-release.markdown | 12 … improving performance for in-memory workloads. We are seeing read QPS as high as 5M (we will writ…
|
| D | 2017-07-25-rocksdb-5-6-1-released.markdown | 9 …ckground_flushes=0`. Instead, users can achieve this by configuring their high-pri thread pool to …
|
| D | 2016-09-28-rocksdb-4-11-2-released.markdown | 47 …s in foreground when destroying iterators. Instead, a job is scheduled in high priority queue and …
|
| D | 2017-12-18-17-auto-tuned-rate-limiter.markdown | 16 …e ratio of intervals where the rate limiter is drained. There are low and high watermarks for this…
|
| D | 2017-08-25-flushwal.markdown | 24 …is makes low commit latency in the storage engine essential for acheiving high throughput. The com…
|
| D | 2014-05-14-lock.markdown | 27 … increased. One such operation is not expensive but if you are building a high throughput server w…
49 … a way that table handler evictions should never happen. It is common for high-throughput, low-lat…
|
| D | 2014-06-27-avoid-expensive-locks-in-get.markdown | 59 …ecting threads to check against a global version which does not change at high frequency, and is t…
|
| D | 2015-10-27-getthreadlist.markdown | 174 // The operation (high-level action) that the current thread is involved.
|
| D | 2015-07-17-spatial-indexing-in-rocksdb.markdown | 14 …s not that big. Earth's entire map data can fit in memory on a reasonably high-end machine. Thus, …
|
| D | 2017-06-26-17-level-based-changes.markdown | 32 We can now see the high-level problems with this approach more clearly. First, locked files in L0 o…
|
| /rocksdb-6.9/db_stress_tool/ |
| D | db_stress_common.cc | 65 int64_t low = 1, mid, high = zipf_sum_size, zipf = 0; in GetOneHotKeyID() local
66 while (low <= high) { in GetOneHotKeyID()
67 mid = (low + high) / 2; in GetOneHotKeyID()
72 high = mid - 1; in GetOneHotKeyID()
|
| /rocksdb-6.9/ |
| D | USERS.md | 9 4. Laser -- Laser is a high query throughput, low (millisecond) latency, key-value storage service …
61 …com/) uses RocksDB as the storage layer for their core key-value storage, high-performance counter…
70 [quasardb](https://www.quasardb.net) is a high-performance, distributed, transactional key-value da…
77 [TiKV](https://github.com/pingcap/tikv) is a GEO-replicated, high-performance, distributed, transac…
83 …/dgraph-io/dgraph) is an open-source, scalable, distributed, low latency, high throughput Graph da…
|
| D | DEFAULT_OPTIONS_HISTORY.md | 14 …ers set higher max_background_compactions but the write throughput is not high, the writes are les…
|
| D | HISTORY.md | 108 …this filter. This replacement is faster and more accurate, especially for high bits per key or mil…
616 …ckground_flushes=0`. Instead, users can achieve this by configuring their high-pri thread pool to …
758 …s in foreground when destroying iterators. Instead, a job is scheduled in high priority queue and …
998 …it. Similarly, options.max_background_flushes is larger than the existing high pri pool of options…
|
| /rocksdb-6.9/docs/_data/ |
| D | features.yml | 8 …w latency storage such as flash drives and high-speed disk drives. RocksDB exploits the full poten…
|
| /rocksdb-6.9/table/plain/ |
| D | plain_table_reader.cc | 453 uint32_t high = upper_bound; in GetOffset() local
461 while (high - low > 1) { in GetOffset()
462 uint32_t mid = (high + low) / 2; in GetOffset()
480 high = mid; in GetOffset()
|
| /rocksdb-6.9/monitoring/ |
| D | histogram_test.cc | 23 uint64_t low, uint64_t high, uint64_t loop = 1) { in PopulateHistogram() argument
25 for (uint64_t i = low; i <= high; i++) { in PopulateHistogram()
|
| /rocksdb-6.9/table/block_based/ |
| D | filter_policy.cc | 358 int high = static_cast<int>(bytes * 8 / bits_per_key_ + 1); in CalculateNumEntry() local
360 int n = high; in CalculateNumEntry()
366 assert(n < high); // High should be an overestimation in CalculateNumEntry()
|
| /rocksdb-6.9/docs/_docs/ |
| D | faq.md | 32 * A message-queue that supports a high number of inserts and deletes.
|
| /rocksdb-6.9/db/ |
| D | db_with_timestamp_basic_test.cc | 129 std::string Timestamp(uint64_t low, uint64_t high) { in Timestamp() argument
132 PutFixed64(&ts, high); in Timestamp()
|
| /rocksdb-6.9/table/ |
| D | table_test.cc | 1044 static bool Between(uint64_t val, uint64_t low, uint64_t high) { in Between() argument
1045 bool result = (val >= low) && (val <= high); in Between()
1050 (unsigned long long)(high)); in Between()
|