Proprietary Database Compression for Fast Access
|
|
Our Services
Xtreme Compression, Inc. is an engineering consultancy offering highly specialized software design expertise. We provide proprietary algorithms to compress structured discrete data for fast access. We apply our unique set of methods, called Xtreme Compression, wherever requirements for size and/or speed demand results beyond the reach of conventional methods.
Results Make Us Unique
Our methods differ radically from convention. They break new technical ground on many fronts. They feature attribute vector coding, the first vector transform method for compressing multidimensional database tables. Unlike conventional methods, attribute vector coding treats databases as databases by modeling data at high levels of abstraction. It systematically exploits multidimensional table structure, takes full advantage of data semantics, and accommodates heterogeneous data types. That makes attribute vector coding far less subject than conventional methods to compressibility limits imposed by information theory.
With Xtreme Compression, data structures are individually compressed and then used as compressed. Decompression is encapsulated and invisible above the lowest software levels. That sets a distinct level above which the host system remains unaffected and unchanged.
The results? Absolute transparency, fast decompression, and compression ratios beyond the reach of conventional methods.
Target Applications
We target large files of relatively static structured data: multidimensional databases with read-intensive query workloads, such as in-memory DBMSs, data warehousing, business intelligence, data mining, statistical databases, analytics, and the like. Xtreme Compression also benefits devices and infrastructures constrained by bandwidth, memory, and/or power dissipation.
Advantages
Xtreme Compression's advantages over conventional methods are vastly superior compression ratio and database access speed. Compression ratios can often be achieved beyond thresholds at which otherwise-infeasible operational goals, products, and services become realities.
|
Aside from Xtreme Compression's reduction in data storage cost, another valuable benefit is faster query processing. Our proprietary methods achieve that by doing the following:
-
Concentrating more information into every physical word read from memory and processed by the CPU, reducing the influence of the ‘von Neumann bottleneck’
-
Storing more real information in internal caches during query to improve data locality and cache hit rates, especially with newer processor technologies
-
Increasing effective disk cache size to retrieve more real information from hard disk per read operation
-
Allowing storage on faster media (i.e., RAM vs hard disk, hard disk vs CD or DVD, local hard disk vs network, etc.)
-
Permitting more access data structures and paths for a given amount of storage
-
Reducing transmission time over links, networks, and other infrastructure components that are expensive to scale
-
In some applications, reducing collision frequency by setting hash table load factors at lower values than what would otherwise be practical
A third benefit for some applications is reduced power consumption from reduced processor cycles.
|
|