Lindsay Powers - The HDF Group The HDF Group provides free, open-source software that is widely used in government, academia and industry. The goal of The HDF Group is to ensure the sustainable development of HDF (Hierarchical Data Format) technologies and the ongoing accessibility of HDF-stored data because users and organizations have mission-critical systems and archives relying on these technologies. These users and organizations are a critical element of the HDF community and an important source of new and innovative uses of, and sustainability for, the HDF platforms, libraries and tools. We want to create a sustainability model for the open access platforms and libraries that can serve these diverse communities in the future use and preservation of their data. As a...

* With Python’s Help

Gerd Heber, The HDF Group

Before the recent release of our PyHexad Excel add-in for HDF5[1], the title might have sounded like the slogan of a global coffee and baked goods chain. That was then. Today, it is an expression of hope for the spreadsheet users who run this country and who either felt neglected by the HDF5 community or who might suffer from a medical condition known as data-bulging workbook stress disorder. In this article, I would like to give you a quick overview of the novel PyHexad therapy and invite you to get involved (after consulting with your doctor).

To access the data in HDF5 files from Excel is a frontrunner among the all-time TOP 10 most frequently asked for features. A spreadsheet tool might be a convenient window into, and user interface for, certain data stored in HDF5 files. Such a tool could help overcome Excel storage and performance limitations, and allow data to be freely “shuttled” between worksheets and HDF5 data containers. PyHexad ([4],[5],[6],[7]) is an attempt to further explore this concept.  

Elena Pourmal, The HDF Group What happened to my compression? One of the most powerful features of HDF5 is the ability to compress or otherwise modify, or “filter,” your data during I/O. By far, the most common user-defined filters are ones that perform data compression.  As you know, there are many compression options. There are filters provided by the HDF5 library (“predefined filters,”) which include several types of filters for data compression, data shuffling and checksum. Users can implement their own “user-defined filters” and employ them with the HDF5 library. [caption id="attachment_10741" align="alignright" width="300"] Cars in a 1973 Philadelphia junkyard – image from National Archives and Records Administration[/caption] While the programming model and usage of the compression filters is straightforward, it is possible for...

...we focus on how far we can push our personal computing devices with Spark. It consists of 7,850 HDF-EOS5 files covering 27 years and totals about 120 GB. We use a driver script, which reads a dataset of interest from each file in the collection, computes per-file quantities of interest, and gathers them in a CSV file for visualization. The processing time on our reference tablet machine for 3.5 years of data using 4 logical processors was about 10 seconds....

Mohamad Chaarawi, The HDF Group

First in a series: parallel HDF5

What costs applications a lot of time and resources rather than doing actual computation?  Slow I/O.  It is well known that I/O subsystems are very slow compared to other parts of a computing system.  Applications use I/O to store simulation output for future use by analysis applications, to checkpoint application memory to guard against system failure, to exercise out-of-core techniques for data that does not fit in a processor’s memory, and so on.  I/O middleware libraries, such as HDF5, provide application users with a rich interface for I/O access to organize their data and store it efficiently.  A lot of effort is invested by such I/O libraries to reduce or completely hide the cost of I/O from applications.

Parallel I/O is one technique used to access data on disk simultaneously from different application processes to maximize bandwidth and speed things up. There are several ways to do parallel I/O, and I will highlight the most popular methods that are in use today.

Blue Waters supercomputer at the National Center for Supercomputing Applications, University of Illinois, Urbana-Champaign campus. Blue Waters is supported by the National Science Foundation and the University of Illinois.

First, to leverage parallel I/O, it is very important that you have a parallel file system;