# psort, Yet Another Fast Stable Sorting Software

@inproceedings{Bertasi2009psortYA, title={psort, Yet Another Fast Stable Sorting Software}, author={Paolo Bertasi and Marco Bressan and Enoch Peserico}, booktitle={SEA}, year={2009} }

psort was the fastest sorting software in 2008 according to the Pennysort benchmark, sorting 181GB of data for 0.01$ of computer time. This paper details its internals, and the careful fitting of its architecture to the structure of modern PCs-class platforms, allowing it to outperform state-of-the-art sorting software such as GNUsort or STXXL .

## Figures and Topics from this paper

## 12 Citations

psort 2009

- 2009

This memo reports the results of our psort (general purpose) sorting software on a 5 disk 427.86$ PC with 4 GB of RAM and an Athlon LE 1640 2.6 GHz processor. psort sorted 67 · 2 records in 2211…

psort 2011 – pennysort , datamation , joulesort ∗

- 2011

This memo reports the results of our psort (general purpose) sorting software on a number of hardware configurations. “Vanilla” psort sorted 10GB for 2122 joules on a Nokia N900 smartphone with an…

Datamation: A Quarter of a Century and Four Orders of Magnitude Later

- Computer Science2011 IEEE International Conference on Cluster Computing
- 2011

Of the many implementation and configuration choices the authors faced, the most crucial were judicious data placement and access patterns on disk, adoption of UDP sockets instead of MPI, careful pruning of virtually all system daemons, and rejection of ``on demand'' frequency scaling.

Energy-efficient sorting using solid state disks

- Computer Science, ChemistryInternational Conference on Green Computing
- 2010

Using a low-power processor, solid state disks, and efficient algorithms, this work beats the current records in the JouleSort benchmark for 10GB to 1 TB of data by factors of up to 5.1.

Energy-efficient sorting using solid state disks

- Computer ScienceSustain. Comput. Informatics Syst.
- 2011

Drawing conclusions on how to design scalable energy-efficient systems for processing large data sets, such as combining relatively weak computing power with high bandwidth storage devices, are led.

Critical Evaluation of Existing External Sorting Methods in the Perspective of Modern Hardware

- Computer ScienceDATESO
- 2015

In this work, original assumptions of the external sorting algorithms are critically evaluated in empirical manner and possible improvements are proposed.

PARALLEL COMPUTING IN ESTIMATION OF PARAMETERS OF ALPHA-STABLE DISTRIBUTION

- Computer Science
- 2012

This paper deals with parallelization problem of alpha-stable dis- tributed data series and estimates the stable parameters of financial data series using the Zolotariov integral presentation of the probability density function with 96-point Gaussian quadrature.

Attacking the tav-128 hash function

- Computer Science
- 2012

This article shows that Tav-128 is not collision resistant and shows that the concatenation of nonlinear functions A and B produces a 64-bit permutation from 32-bit messages which could be a useful light weight primitive for future RFID protocols.

Cryptanalysis of Tav-128 Hash Function

- Computer ScienceINDOCRYPT
- 2010

This article shows a practical collision attack on Tav-128 having a complexity of 237 calls to the compression function and produces message pairs of arbitrary length which produce the same hash value under this hash function.

Revize metod externího třídění pro moderní hardware

- Physics, Computer ScienceITAT
- 2013

Dale p�’edkladame empiricke srovnani s již existujicimi metodami nový přistup k externimu třiděni, který reflektuje parametry soucasneho hardware.

## References

SHOWING 1-10 OF 30 REFERENCES

SheenkSort: 2003 Performance / Price Sort and PennySort

- Computer Science
- 2003

This paper presents the main considerations for SheenkSort and reports the results for PennySort, Performance / Price Sort, as well as Datamation Sort and Minute Sort and the YHSort Framework.

Performance / Price Sort

- Computer ScienceArXiv
- 1998

This paper documents this and proposes that the PennySort benchmark be revised to Performance/Price sort: a simple GB/$ sort metric based on a two-pass external sort.

Performance / Price Sort and PennySort

- Computer Science
- 1998

This paper documents this and proposes that the PennySort benchmark be revised to Performance/Price sort: a simple GB/$ sort metric based on a two-pass external sort.

Nsort: a Parallel Sorting Program for NUMA and SMP Machines

- Computer Science
- 2000

Ordinal TM Nsort TM is a high-performance sort program for SGI IRIX, Sun Solaris and HP-UX servers that can use tens of processors and hundreds of disks to quickly sort and merge data.

High-performance sorting on networks of workstations

- Computer ScienceSIGMOD '97
- 1997

We report the performance of NOW-Sort, a collection of sorting implementations on a Network of Workstations (NOW). We find that parallel sorting on a NOW is competitive to sorting on the large-scale…

Parallel sorting on a shared-nothing architecture using probabilistic splitting

- Computer Science[1991] Proceedings of the First International Conference on Parallel and Distributed Information Systems
- 1991

The authors consider the problem of external sorting in a shared-nothing multiprocessor with two techniques for determining ranges of sort keys: exact splitting, using a parallel version of the algorithm proposed by Iyer, Ricard, and Varman; and probabilistic splitting, which uses sampling to estimate quantiles.

Alphasort: A cache-sensitive parallel external sort

- Computer ScienceThe VLDB Journal
- 2005

A new sort algorithm, called AlphaSort, demonstrates that commodity processors and disks can handle commercial batch workloads and argues that modern architectures require algorithm designers to re-examine their use of the memory hierarchy.

A "Measure of Transaction Processing" 20 Years Later

- Computer ScienceIEEE Data Eng. Bull.
- 2005

It is shown that improvement has exceeded Moore’s law – largely due to hardware improvements, software improvements, massive parallelism, and changing from mainframe to commodity economics.

GPUTeraSort: high performance graphics co-processor sorting for large database management

- Computer ScienceSIGMOD Conference
- 2006

Overall, the results indicate that using a GPU as a co-processor can significantly improve the performance of sorting algorithms on large databases.

Asynchronous parallel disk sorting

- Computer ScienceSPAA '03
- 2003

We develop an algorithm for parallel disk sorting, whose I/O cost approaches the lower bound and that guarantees almost perfect overlap between I/O and computation. Previous algorithms have either…