An official website of the United States government
In this paper, we present our GPU implementation of the quotient filter, a compact data structure designed to implement approximate membership queries. The quotient filter is similar to the more well-known Bloom filter; however, in addition to set insertion and membership queries, the quotient filter also supports deletions and merging filters without requiring rehashing of the data set. Furthermore, the quotient filter can be extended to include counters without increasing the memory footprint. This paper describes our GPU implementation of two types of quotient filters: the standard quotient filter and the rank-and-select-based quotient filter. We describe the parallelization of all filter operations, including a comparison of the four different methods we devised for parallelizing quotient filter construction. In solving this problem, we found that we needed an operation similar to a parallel scan, but for non-associative operators. One outcome of this work is a variety of methods for computing parallel scan-type operations on a non-associative operator. For membership queries, we achieve a throughput of up to 1.13 billion items/second for the rank-and-select-based quotient filter: a speedup of 3x over the BloomGPU filter. Our fastest filter build method achieves a speedup of 2.1-3.1x over BloomGPU, with a peak throughput of 621 million items/second, and a rate of 516 million items/second for a 70% full filter. However, we find that our filters do not perform incremental updates as fast as the BloomGPU filter. For a batch of 2 million items, we perform incremental inserts at a rate of 81 million items/second - a 2.5x slowdown compared to BloomGPU's throughput of 201 million items/second. The quotient filter's memory footprint is comparable to that of a Bloom filter.
more »
« less
Vector Quotient Filters: Overcoming the Time/Space Trade-Off in Filter Design
Today’s filters, such as quotient, cuckoo, and Morton, have a trade-off between space and speed; even when moderately full (e.g., 50%-75% full), their performance degrades nontrivially. The result is that today’s systems designers are forced to choose between speed and space usage. In this paper, we present the vector quotient filter (VQF). Locally, the VQF is based on Robin Hood hashing, like the quotient filter, but uses power-of-two-choices hashing to reduce the variance of runs, and thus others consistent, high throughput across load factors. Power-of-two-choices hashing also makes it more amenable to concurrent updates, compared to the cuckoo filter and variants. Finally, the vector quotient filter is designed to exploit SIMD instructions so that all operations have $ (1) cost, independent of the size of the filter or its load factor. We show that the vector quotient flter is 2x faster for inserts compared to the Morton filter (a cuckoo filter variant and state-of- the-art for inserts) and has similar lookup and deletion performance as the cuckoo filter (which is fastest for queries and deletes), despite having a simpler design and implementation. The vector quotient filter has minimal performance decline at high load factors, a problem that has plagued modern filters, including quotient, cuckoo, and Morton. Furthermore, we give a thread-safe version of the vector quotient filter and show that insertion throughput scales 3x with four threads compared to a single thread.
more »
« less
- PAR ID:
- 10298533
- Date Published:
- Journal Name:
- SIGMOD record
- ISSN:
- 1943-5835
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
In this paper, we present our GPU implementation of the quotient filter, a compact data structure designed to implement approximate membership queries. The quotient filter is similar to the more well-known Bloom filter; however, in addition to set insertion and membership queries, the quotient filter also supports deletions and merging filters without requiring rehashing of the data set. Furthermore, the quotient filter can be extended to include counters without increasing the memory footprint. This paper describes our GPU implementation of two types of quotient filters: the standard quotient filter and the rank-and-select-based quotient filter. We describe the parallelization of all filter operations, including a comparison of the four different methods we devised for parallelizing quotient filter construction. In solving this problem, we found that we needed an operation similar to a parallel scan, but for non-associative operators. One outcome of this work is a variety of methods for computing parallel scan-type operations on a non-associative operator. For membership queries, we achieve a throughput of up to 1.13 billion items/second for the rank-and-select-based quotient filter: a speedup of 3x over the BloomGPU filter. Our fastest filter build method achieves a speedup of 2.1--3.1x over BloomGPU, with a peak throughput of 621 million items/second, and a rate of 516 million items/second for a 70% full filter. However, we find that our filters do not perform incremental updates as fast as the BloomGPU filter. For a batch of 2 million items, we perform incremental inserts at a rate of 81 million items/second -- a 2.5x slowdown compared to BloomGPU's throughput of 201 million items/second. The quotient filter's memory footprint is comparable to that of a Bloom filter.more » « less
-
null (Ed.)RDMA has been an important building block for many high-performance distributed key-value stores in recent prior work. To sustain millions of operations per second per node, many of these works use hashing schemes, such as cuckoo hashing, that guarantee that an existing key can be found in a small, constant number of RDMA operations. In this paper, we explore whether linear probing is a compelling design alternative to cuckoo-based hashing schemes. Instead of reading a fixed number of bytes per RDMA request, this paper introduces a mathematical model that picks the optimal read size by balancing the cost of performing an RDMA operation with the probability of encountering a probe sequence of a certain length. The model can further leverage optimization hints about the location of clusters of keys, which commonly occur in skewed key distributions. We extensively evaluate the performance of linear probing with a variable read size in a modern cluster to understand the trade-offs between cuckoo hashing and linear probing. We find that cuckoo hashing outperforms linear probing only in very highly loaded hash tables (load factors at 90% or higher) that would be prohibitively expensive to maintain in practice. Overall, linear probing with variable-sized reads using our model has up to 2.8× higher throughput than cuckoo hashing, with throughput as high as 50M lookup operations per second per node.more » « less
-
Filters trade off accuracy for space and occasionally return false positive matches with a bounded error. Numerous systems use filters in fast memory to avoid performing expensive I/Os to slow storage. A fundamental limitation in traditional filters is that they do not change their representation upon seeing a false positive match. Therefore, the maximum false positive rate is only guaranteed for a single query, not for an arbitrary set of queries. We can improve the filter's performance on a stream of queries, especially on a skewed distribution, if we can adapt after encountering false positives. Adaptive filters, such as telescoping quotient filters and adaptive cuckoo filters, update their representation upon detecting a false positive to avoid repeating the same error in the future. Adaptive filters require an auxiliary structure, typically much larger than the main filter and often residing on slow storage, to facilitate adaptation. However, existing adaptive filters are not practical and have not been adopted in real-world systems for two main reasons. First, they offer weak adaptivity guarantees, meaning that fixing a new false positive can cause a previously fixed false positive to come back. Secondly, the sub-optimal design of the auxiliary structure results in adaptivity overheads so substantial that they can actually diminish overall system performance compared to a traditional filter. In this paper, we design and implement the \sysname, the first practical adaptive filter with minimal adaptivity overhead and strong adaptivity guarantees, which means that the performance and false-positive guarantees continue to hold even for adversarial workloads. The \sysname is based on the state-of-the-art quotient filter design and preserves all the critical features of the quotient filter such as cache efficiency and mergeability. Furthermore, we employ a new auxiliary structure design which results in considerably low adaptivity overhead and makes the \sysname practical in real systems. We evaluate the \sysname by using it to filter queries to an on-disk B-tree database and find no negative impact on insert or query performance compared to traditional filters. Against adversarial workloads, the \sysname preserves system performance, whereas traditional filters incur 2× slowdown from adversaries representing as low as 1% of the workload. Finally, we show that on skewed query workloads, the \sysname can reduce the false-positive rate 100× using negligible (1/1000th of a bit per item) space overhead.more » « less
-
We revisit the problem of building static hash tables on the GPU and present an efficient implementation of bucketed hash tables. By decoupling the probing scheme from the hash table in-memory representation, we offer an implementation where the number of probes and the bucket size are the only factors limiting performance. Our analysis sweeps through the hash table parameter space for two probing schemes: cuckoo and iceberg hashing. We show that a bucketed cuckoo hash table (BCHT) that uses three hash functions outperforms alternative methods that use iceberg hashing and a cuckoo hash table that uses a bucket size of one. At load factors as high as 0.99, BCHT enjoys an average probe count of 1.43 during insertion. Using three hash functions only, positive and negative queries require at most 1.39 and 2.8 average probes per key, respectively.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
