arXiv subjects

Owen Kaser

Publications and source records attributed to Owen Kaser.

At least 19 recordsLinked to original sources

arXiv2022

Roaring Bitmaps: Implementation of an Optimized Software Library

Compressed bitmap indexes are used in systems such as Git or Oracle to accelerate queries. They represent sets and often support operations such as unions, intersections, differences, and symmetric differences. Several important systems such as Elasticsearch, Apache Spark, Netflix's Atlas, LinkedIn's Pinot, Metamarkets' Druid, Pilosa, Apache Hive, Apache Tez, Microsoft Visual Studio Team Services and Apache Kylin rely on a specific type of compressed bitmap index called Roaring. We present an optimized software library written in C implementing Roaring bitmaps: CRoaring. It benefits from several algorithms designed for the single-instruction-multiple-data (SIMD) instructions available on commodity processors. In particular, we present vectorized algorithms to compute the intersection, union, difference and symmetric difference between arrays. We benchmark the library against a wide range of competitive alternatives, identifying weaknesses and strengths in our software. Our work is available under a liberal open-source license.

cs.DB↗

arXiv2021

Integer Division by Constants: Optimal Bounds

The integer division of a numerator n by a divisor d gives a quotient q and a remainder r. Optimizing compilers accelerate software by replacing the division of n by d with the division of c * n (or c * n + c) by m for convenient integers c and m chosen so that they approximate the reciprocal: c/m ~= 1/d. Such techniques are especially advantageous when m is chosen to be a power of two and when d is a constant so that c and m can be precomputed. The literature contains many bounds on the distance between c/m and the divisor d. Some of these bounds are optimally tight, while others are not. We present optimally tight bounds for quotient and remainder computations.

cs.DS↗

arXiv2020

Hierarchical Bin Buffering: Online Local Moments for Dynamic External Memory Arrays

Local moments are used for local regression, to compute statistical measures such as sums, averages, and standard deviations, and to approximate probability distributions. We consider the case where the data source is a very large I/O array of size n and we want to compute the first N local moments, for some constant N. Without precomputation, this requires O(n) time. We develop a sequence of algorithms of increasing sophistication that use precomputation and additional buffer space to speed up queries. The simpler algorithms partition the I/O array into consecutive ranges called bins, and they are applicable not only to local-moment queries, but also to algebraic queries (MAX, AVERAGE, SUM, etc.). With N buffers of size sqrt{n}, time complexity drops to O(sqrt n). A more sophisticated approach uses hierarchical buffering and has a logarithmic time complexity (O(b log_b n)), when using N hierarchical buffers of size n/b. Using Overlapped Bin Buffering, we show that only a single buffer is needed, as with wavelet-based algorithms, but using much less storage. Applications exist in multidimensional and statistical databases over massive data sets, interactive image processing, and visualization.

cs.DS↗

arXiv2019

Faster Remainder by Direct Computation: Applications to Compilers and Software Libraries

On common processors, integer multiplication is many times faster than integer division. Dividing a numerator n by a divisor d is mathematically equivalent to multiplication by the inverse of the divisor (n / d = n x 1/d). If the divisor is known in advance---or if repeated integer divisions will be performed with the same divisor---it can be beneficial to substitute a less costly multiplication for an expensive division. Currently, the remainder of the division by a constant is computed from the quotient by a multiplication and a subtraction. But if just the remainder is desired and the quotient is unneeded, this may be suboptimal. We present a generally applicable algorithm to compute the remainder more directly. Specifically, we use the fractional portion of the product of the numerator and the inverse of the divisor. On this basis, we also present a new, simpler divisibility algorithm to detect nonzero remainders. We also derive new tight bounds on the precision required when representing the inverse of the divisor. Furthermore, we present simple C implementations that beat the optimized code produced by state-of-art C compilers on recent x64 processors (e.g., Intel Skylake and AMD Ryzen), sometimes by more than 25%. On all tested platforms including 64-bit ARM and POWER8, our divisibility-test functions are faster than state-of-the-art Granlund-Montgomery divisibility-test functions, sometimes by more than 50%.

cs.MS↗

arXiv2018

Strongly universal string hashing is fast

We present fast strongly universal string hashing families: they can process data at a rate of 0.2 CPU cycle per byte. Maybe surprisingly, we find that these families---though they require a large buffer of random numbers---are often faster than popular hash functions with weaker theoretical guarantees. Moreover, conventional wisdom is that hash functions with fewer multiplications are faster. Yet we find that they may fail to be faster due to operation pipelining. We present experimental results on several processors including low-powered processors. Our tests include hash functions designed for processors with the Carry-Less Multiplication (CLMUL) instruction set. We also prove, using accessible proofs, the strong universality of our families.

cs.DB↗

arXiv2018

Consistently faster and smaller compressed bitmaps with Roaring

Compressed bitmap indexes are used in databases and search engines. Many bitmap compression techniques have been proposed, almost all relying primarily on run-length encoding (RLE). However, on unsorted data, we can get superior performance with a hybrid compression technique that uses both uncompressed bitmaps and packed arrays inside a two-level tree. An instance of this technique, Roaring, has recently been proposed. Due to its good performance, it has been adopted by several production platforms (e.g., Apache Lucene, Apache Spark, Apache Kylin and Druid). Yet there are cases where run-length encoded bitmaps are smaller than the original Roaring bitmaps---typically when the data is sorted so that the bitmaps contain long compressible runs. To better handle these cases, we build a new Roaring hybrid that combines uncompressed bitmaps, packed arrays and RLE compressed segments. The result is a new Roaring format that compresses better. Overall, our new implementation of Roaring can be several times faster (up to two orders of magnitude) than the implementations of traditional RLE-based alternatives (WAH, Concise, EWAH) while compressing better. We review the design choices and optimizations that make these good results possible.

cs.DB↗

arXiv2016

Threshold and Symmetric Functions over Bitmaps

Bitmap indexes are routinely used to speed up simple aggregate queries in databases. Set operations such as intersections, unions and complements can be represented as logical operations (AND, OR, NOT). However, less is known about the application of bitmap indexes to more advanced queries. We want to extend the applicability of bitmap indexes. As a starting point, we consider symmetric Boolean queries (e.g., threshold functions). For example, we might consider stores as sets of products, and ask for products that are on sale in 2 to 10 stores. Such symmetric Boolean queries generalize intersection, union, and T-occurrence queries. It may not be immediately obvious to an engineer how to use bitmap indexes for symmetric Boolean queries. Yet, maybe surprisingly, we find that the best of our bitmap-based algorithms are competitive with the state-of-the-art algorithms for important special cases (e.g., MergeOpt, MergeSkip, DivideSkip, ScanCount). Moreover, unlike the competing algorithms, the result of our computation is again a bitmap which can be further processed within a bitmap index. We review algorithmic design issues such as the aggregation of many compressed bitmaps. We conclude with a discussion on other advanced queries that bitmap indexes might be able to support efficiently.

cs.DB↗

arXiv2016

Removing Manually-Generated Boilerplate from Electronic Texts: Experiments with Project Gutenberg e-Books

Collaborative work on unstructured or semi-structured documents, such as in literature corpora or source code, often involves agreed upon templates containing metadata. These templates are not consistent across users and over time. Rule-based parsing of these templates is expensive to maintain and tends to fail as new documents are added. Statistical techniques based on frequent occurrences have the potential to identify automatically a large fraction of the templates, thus reducing the burden on the programmers. We investigate the case of the Project Gutenberg corpus, where most documents are in ASCII format with preambles and epilogues that are often copied and pasted or manually typed. We show that a statistical approach can solve most cases though some documents require knowledge of English. We also survey various technical solutions that make our approach applicable to large data sets.

cs.DL↗

arXiv2016

Sorting improves word-aligned bitmap indexes

Bitmap indexes must be compressed to reduce input/output costs and minimize CPU usage. To accelerate logical operations (AND, OR, XOR) over bitmaps, we use techniques based on run-length encoding (RLE), such as Word-Aligned Hybrid (WAH) compression. These techniques are sensitive to the order of the rows: a simple lexicographical sort can divide the index size by 9 and make indexes several times faster. We investigate row-reordering heuristics. Simply permuting the columns of the table can increase the sorting efficiency by 40%. Secondary contributions include efficient algorithms to construct and aggregate bitmaps. The effect of word length is also reviewed by constructing 16-bit, 32-bit and 64-bit indexes. Using 64-bit CPUs, we find that 64-bit indexes are slightly faster than 32-bit indexes despite being nearly twice as large.

cs.DB↗

arXiv2016

Recursive n-gram hashing is pairwise independent, at best

Many applications use sequences of n consecutive symbols (n-grams). Hashing these n-grams can be a performance bottleneck. For more speed, recursive hash families compute hash values by updating previous values. We prove that recursive hash families cannot be more than pairwise independent. While hashing by irreducible polynomials is pairwise independent, our implementations either run in time O(n) or use an exponential amount of memory. As a more scalable alternative, we make hashing by cyclic polynomials pairwise independent by ignoring n-1 bits. Experimentally, we show that hashing by cyclic polynomials is is twice as fast as hashing by irreducible polynomials. We also show that randomized Karp-Rabin hash families are not pairwise independent.

cs.DB↗

arXiv2016

Better bitmap performance with Roaring bitmaps

Bitmap indexes are commonly used in databases and search engines. By exploiting bit-level parallelism, they can significantly accelerate queries. However, they can use much memory, and thus we might prefer compressed bitmap indexes. Following Oracle's lead, bitmaps are often compressed using run-length encoding (RLE). Building on prior work, we introduce the Roaring compressed bitmap format: it uses packed arrays for compression instead of RLE. We compare it to two high-performance RLE-based bitmap encoding techniques: WAH (Word Aligned Hybrid compression scheme) and Concise (Compressed `n' Composable Integer Set). On synthetic and real data, we find that Roaring bitmaps (1) often compress significantly better (e.g., 2 times) and (2) are faster than the compressed alternatives (up to 900 times faster for intersections). Our results challenge the view that RLE-based bitmap compression is best.

cs.DB↗

arXiv2015

Faster 64-bit universal hashing using carry-less multiplications

Intel and AMD support the Carry-less Multiplication (CLMUL) instruction set in their x64 processors. We use CLMUL to implement an almost universal 64-bit hash family (CLHASH). We compare this new family with what might be the fastest almost universal family on x64 processors (VHASH). We find that CLHASH is at least 60% faster. We also compare CLHASH with a popular hash function designed for speed (Google's CityHash). We find that CLHASH is 40% faster than CityHash on inputs larger than 64 bytes and just as fast otherwise.

cs.DS↗

arXiv2014

Compressed bitmap indexes: beyond unions and intersections

Compressed bitmap indexes are used to speed up simple aggregate queries in databases. Indeed, set operations like intersections, unions and complements can be represented as logical operations (AND,OR,NOT) that are ideally suited for bitmaps. However, it is less obvious how to apply bitmaps to more advanced queries. For example, we might seek products in a store that meet some, but maybe not all, criteria. Such threshold queries generalize intersections and unions; they are often used in information-retrieval and data-mining applications. We introduce new algorithms that are sometimes three orders of magnitude faster than a naive approach. Our work shows that bitmap indexes are more broadly applicable than is commonly believed.

cs.DB↗

arXiv2014

Diamond Dicing

In OLAP, analysts often select an interesting sample of the data. For example, an analyst might focus on products bringing revenues of at least 100 000 dollars, or on shops having sales greater than 400 000 dollars. However, current systems do not allow the application of both of these thresholds simultaneously, selecting products and shops satisfying both thresholds. For such purposes, we introduce the diamond cube operator, filling a gap among existing data warehouse operations. Because of the interaction between dimensions the computation of diamond cubes is challenging. We compare and test various algorithms on large data sets of more than 100 million facts. We find that while it is possible to implement diamonds in SQL, it is inefficient. Indeed, our custom implementation can be a hundred times faster than popular database engines (including a row-store and a column-store).

cs.DB↗

arXiv2014

One-Pass, One-Hash n-Gram Statistics Estimation

In multimedia, text or bioinformatics databases, applications query sequences of n consecutive symbols called n-grams. Estimating the number of distinct n-grams is a view-size estimation problem. While view sizes can be estimated by sampling under statistical assumptions, we desire an unassuming algorithm with universally valid accuracy bounds. Most related work has focused on repeatedly hashing the data, which is prohibitive for large data sources. We prove that a one-pass one-hash algorithm is sufficient for accurate estimates if the hashing is sufficiently independent. To reduce costs further, we investigate recursive random hashing algorithms and show that they are sufficiently independent in practice. We compare our running times with exact counts using suffix arrays and show that, while we use hardly any storage, we are an order of magnitude faster. The approach further is extended to a one-pass/one-hash computation of n-gram entropy and iceberg counts. The experiments use a large collection of English text from the Gutenberg Project as well as synthetic data.

cs.DB↗

arXiv2014

Reordering Rows for Better Compression: Beyond the Lexicographic Order

Sorting database tables before compressing them improves the compression rate. Can we do better than the lexicographical order? For minimizing the number of runs in a run-length encoding compression scheme, the best approaches to row-ordering are derived from traveling salesman heuristics, although there is a significant trade-off between running time and compression. A new heuristic, Multiple Lists, which is a variant on Nearest Neighbor that trades off compression for a major running-time speedup, is a good option for very large tables. However, for some compression schemes, it is more important to generate long runs rather than few runs. For this case, another novel heuristic, Vortex, is promising. We find that we can improve run-length encoding up to a factor of 3 whereas we can improve prefix coding by up to 80%: these gains are on top of the gains due to lexicographically sorting the table. We prove that the new row reordering is optimal (within 10%) at minimizing the runs of identical values within columns, in a few cases.

cs.DB↗

arXiv2011

Reordering Columns for Smaller Indexes

Column-oriented indexes-such as projection or bitmap indexes-are compressed by run-length encoding to reduce storage and increase speed. Sorting the tables improves compression. On realistic data sets, permuting the columns in the right order before sorting can reduce the number of runs by a factor of two or more. Unfortunately, determining the best column order is NP-hard. For many cases, we prove that the number of runs in table columns is minimized if we sort columns by increasing cardinality. Experimentally, sorting based on Hilbert space-filling curves is poor at minimizing the number of runs.

cs.DB↗

arXiv2009

Histogram-Aware Sorting for Enhanced Word-Aligned Compression in Bitmap Indexes

Bitmap indexes must be compressed to reduce input/output costs and minimize CPU usage. To accelerate logical operations (AND, OR, XOR) over bitmaps, we use techniques based on run-length encoding (RLE), such as Word-Aligned Hybrid (WAH) compression. These techniques are sensitive to the order of the rows: a simple lexicographical sort can divide the index size by 9 and make indexes several times faster. We investigate reordering heuristics based on computed attribute-value histograms. Simply permuting the columns of the table based on these histograms can increase the sorting efficiency by 40%.

cs.DB↗