An Experimental Testbed for Data-Intensive Systems and Applications
We live in a world populated by enormous and ever-growing amounts of data from a wide variety of sources such as the Web, social networks, satellite telemetry, sensors (e.g., medical, environmental, and agricultural), and networked services. There is a compelling human need (1) to represent, analyze, query, manage, understand, and respond to such data for knowledge extraction and decision making, and (2) to build better operating and distributed systems that can support applications and parallel programming that deals with such large-scale data.
What's Unique Here?
The Illinois Cloud Computing Testbed (CCT) is the world's first cloud testbed aimed at supporting both systems innovation and applications research within a single microcosm.
CCT is unique in several respects: (1) it is a true cloud/datacenter testbed, e.g., its storage to computation ratio is different from that of many existing testbeds such as Emulab and PlanetLab). Currently, CCT is configured with about 500 TB of shared storage and 1000+ shared cores; (2) it is the only cloud testbed to support both applications and systems research (in contrast with the Google-IBM testbed, for example).
Current research efforts on CCT, both internal to the University of Illinois and external, include but are not limited to the following research areas: networking, operating systems, databases, storage, virtual machines, distributed systems, data-mining, Web search, network measurements, and multimedia.
CCT is currently running NSF-funded projects as well as projects from the Department of Computer Science and NCSA at the University of Illinois.
Member of the First Global Cloud Testbed
CCT is one of six sites participating in the world's first networked cloud testbed, called Open Cirrus. Other participating member sites include Intel, Yahoo!, HP, the Infocomm Development Authority (IDA) of Singapore, and the Karlsruhe Institute of Technology in Germany. The number of sites, as well as the size of each site within OpenCirrus, is growing rapidly. OpenCirrus is aimed towards enabling systems and applications research for data-intensive computing that spans multi-site clouds.
How Do I Access CCT?
CCT is open to (1) external researchers outside the University of Illinois (e.g., NSF's Data-Intensive Computing program) who are provided access by NSF (please see NSF's instructions for requesting access to CCT), and (2) internal University of Illinois researchers, both within and outside the Department of Computer Science. For more information visit the User Allocations page.
The CCT infrastructure can be accessed in one of the following two modes:
- Systems Access: Researchers are allowed low-level access to systems resources on the CCT machines. By default, machines are currently configured with CentOS (the underlying OS beneath many popular parallel programming frameworks such as Hadoop and Pig Latin), and users receive dedicated sudo access to a small set of machines. Thus, CCT enables researchers to innovate deep inside the system software stack, supporting exploration and creation of novel system-level support for data-intensive computing. Further research opportunities accrue from CCT's link to the OpenCirrus global testbed, thereby enabling research on both single cloud and federated/meshed cloud environments.
- Hadoop/Pig Service: CCT also supports applications research using stable parallel programming services. Hadoop and Pig Latin are currently available on CCT and are being used by many researchers for data-intensive applications.
For more information and to get started, visit the User Allocations page.
CCT was created from resources provided jointly by the National Science Foundation (NSF), Yahoo!, Intel, Hewlett-Packard, and the University of Illinois at Urbana-Champaign. CCT is housed within and supported primarily by staff and faculty from the Department of Computer Science at the University of Illinois.