[11/12/2013] Dr. Valliappa Lakshmanan is visitting CCNY from CIMMS at U. of Oklahoma and NOAA NSSL [Talk Flyer]
[08/01/2013] We won a 4-year NSF award (2013-2017) totaling $449,845 entitled "Spatial Data and Trajectory Data Management on GPUs" [NSF Award Page][07/07/2014] The first year report of our NSF project was apporved by NSF. We are looking forward to a productive second year.
[07/08/2014] I visited IBM T.J. Watson Resesarch Center and gave a talk entiled "High-Performance Processing of Large-Scale Geospatial Data on GPUs" to the IBM System G group.[07/10/2014] Two high school students, Rabia Akhta (Stuyvesant High School) and Juliana Sempertegui (City College Academy of the Arts), together with their graduate student mentor Jose Pillich, joined the GeoTECI lab this summer through NOAA-CREST Summer High School Internship Program (SHIP) at CCNY. Welcome!
[08/21/2014] I visited Pitney Bowes (PB) Strategic Technology and Innovation Center and gave a talk entiled "Large-Scale Spatial Data Management on New Hardware: A Case Study Using Taxi Trip Data in NYC". PB is the parent company of MapInfo, a major GIS vendor. I have been using Mapinfo since version 4.0 back to early 1990s - great pleasure!
[08/22/2014] Our paper, entitled "Data Parallel Quadtree Indexing and Spatial Query Processing of Complex Polygon Data on GPUs" was accepted by the Fifth International Workshop on Accelerating Data Management Systems Using Modern Processor and Storage Architectures (ADMS'14), collocated with VLDB'14 in Hangzhou, China (09/01-09/05)
[/12/24/2014] Our submission entitled "Large-Scale Spatial Join Query Processing in Cloud " is accepted by ICDE CloudDM'15 workshop. The work provides high-level descriptions and performance comparisons of the SpatialSpark and ISP-MC prototypes that we we have developed. The two systems are targeted for existing Cloud computing resources by using existing mature geometry libraries (JTS and GEOS, respective). The technical report version of the paper is here.
submission entitled "Scalable and Efficient Spatial Data
Management on Multi-Core CPU and GPU Clusters: A Preliminary
Implementation based on Impala" has been accepted by ICDE HardBD'15
workshop. By reusing ISP-MC framework,
developing new AOS (Array of Structures) for geometry layout and
integrating data parallel designs for efficient single-node
compuitng, both ISP-MC+ (for multi-core CPU clusters) and
ISP-GPU (for GPU-accelerated clusters) have achieved impressive
performance. The technical report version of the paper is here.
[04/10/2015] Our paper entitled "Tiny GPU Cluster for Big Spatial Data: A Preliminary Performance Evaluation" was accepted by IEEE International Workshop on High-Performance Big Data Computing (HPBDC'15) workshop, colocated with the main IEEE ICDCS'15 conference. Click here for the PDF file (6 formatted pages- IEEE Conf.).
[04/15/2015] Prof. W. Randolph Franklin from RPI visited our lab. Three talks were given by his research group. His talk, entitled "Algorithms, libraries, and development environments to process huge geoinformatic databases on modern hardware" is online.[05/03/2015] Our submission entitled "A Lightweight Distributed Execution Engine for Large-Scale Spatial Join Query Processing" is accepted by IEEE Big Data Congres 2015. The acceptance rate is 20% and it happens to be at NYC this year. Click here for the PDF file (8 formatted pages- IEEE Conf.).