Page Operations
Browse Space
Ocean Observatories Infrastructure Cyberinfrastructure Prototype Development
This page describes completed CI prototype development activities together with their results. The OOI Cyberinfrastructure prototype development team uses the prototype pages for ongoing collaboration and result documentation. The tables below list the OOI CI prototypes. Prototypes involve many partner organizations and explore the wide range of technologies that are part of the CI design.
Note: Ongoing and planned prototyping during the inception and elaboration phases of releases 1-5 are documented on the respective subsystem development pages.
Compressed Pilot Period Prototypes (Jan-Dec 2009)
During the OOI pilot period (Jan to Dec 2009), the emphasis is on prototype development and community deployments. The main purpose of pilot prototype projects is risk mitigation for the OOI construction period, starting Sept 2009. Many of the projects follow a 6 week iteration cycle. The compression of the pilot period due to the early construction of the OOI lead to a shift in the prototype schedule and the delay of some prototypes.
- Background Materials provides references to selected background materials
- Pilot Prototype Planning and Integration describes overarching planning and design for the prototype projects during the pilot period
- Pilot Period Final Reports: COI, CEI, DM, S&A, PP+A&S (OSSE-Report)
| Prototype Name | Participants | Technical Scope, Comments | Dates |
Status | SoW |
Reports etc. |
|---|---|---|---|---|---|---|
| Agent Contract Network | UCSD, NCSU |
Researching the fundamentals of contracts and commitments in distributed applications spanning multiple domains of authority | Feb 09 to Dec 09 |
completed | SoW | Deliverables, COI-Report |
| Cloud Provisioning Environment | UCSD, ANL |
Deploying and managing cloud computing resources automatically and based on demand | Mar 09 to Oct 09 | completed | SoW |
Poster, CEI-Report |
| Collaboration Tools Cloud Deployment | UCSD, OL |
Deploying the OOI collaboration tools (Confluence, Jira, Alfresco) in the cloud for increased availability | Mar 09 to Jul 09 | completed |
CEI-Report | |
| Data Exchange | UCSD, NOAA (IOOS) |
Developing a cloud deployed science data distribution and processing infrastructure. This infrastructure should serve numerical modeling communities and provides cloud based data set caching, cataloging and processing. Server-side processing options include Ferret, Gridfields and Matlab, covering structured and unstructured gridded data | Feb 09 to Feb 10 |
preliminary prototype available |
SoW |
Poster, DM-Report, see Signell e.a. JMS 2008 |
| Global Load Balancing | UCSD |
Investigating a reliable global network routing strategy based on global site selectors |
Mar 09 to Aug 09 | completed |
|
CEI-Report |
| Messaging Service: Client Adapter | UCSD | Developing an architecture and Python implementation for a framework that adapts any application and service to the Messaging service for reliable communication |
Feb 09 to Oct 09 | completed (continued in COI) | |
COI-Report |
| Messaging Service: Distributed IPC Facility |
UCSD, Infinetics, RabbitMQ | Investigating and developing a fundamental infrastructure for distributed inter-process communication (IPC) that requires enrollment of all participants and explicit allocation of networking resources before any communication, i.e. read/write on distributed information state. | Sep 09 to Dec 09 | completed |
SoW | Final Report, Architecture Report, API-doc, COI-Report |
| Messaging Service: Broker Infrastructure | UCSD, RabbitMQ |
Develop the core transport and communication mechanism of the distributed system, based on a group of AMQP message broker installations. Virtualizes AMQP broker capabilities throughout the distributed system spanning multiple domains of authority. |
Sep 09 to Dec 09 | completed | SoW | Final Report, API-doc, COI-Report |
| OMF Collaboration Prototype | UCSD, MBARI, NCSA, Lindquist |
Integrating RedHat MRG and QMF with Instrument Proxies and Identity Management. Based on the OMF prototype using Enterprise Service Bus technology. Replace ESB with MRG. Provide an instrument agent platform that interfaces with Antelope ORB (via a to-be-developed ORB2AMQP bridge) and with existing SIAM instrument drivers fronting a serial and an Ethernet instrument |
Aug 09 to Dec 09 | completed |
SoW |
SA-Report |
| OSSE - Ocean System Simulation Experiment | UCSD, JPL, MIT, Rutgers | Integrating observational data and forecast model data products for the purpose of tasking autonomous resources, such as gliders. Based on the Data Exchange infrastructure for science data distribution. Applies the NASA JPL tools CASPER and ASPEN for observation resource planning and MIT's MOOS-IvP for autonomous control of mobile assets such as AUVs. The experiments and deployments are carried out in the Mid Atlantic Bight. | Feb 09 to Dec 09 | completed | OSSE Pilot Plan | OSSE-Report |
| OSSE MaRCOOS Model Integration | OSSE team, MaRCOOS | Integrating historical and operational model data products as well as observational data (SST, ocean color, glider profiles) for the OSSE in the MaRCOOS region | Feb 09 to Dec 09 | completed |
OSSE Pilot Plan | OSSE-Report |
| Semantic Framework Prototype | UCSD, MBARI, MMI |
Providing access to semantically enabled science data. Based on the Data Exchange prototype and technologies. Uses OpenDAP and the VSTO semantic framework and applies MMI (Marine Metadata Initiative) vocabularies and tools. | Sep 09 to Dec 09 | completed |
SoW |
DM-Report |
| SQLstream Prototype | UCSD, SQLstream |
Evaluating the SQLstream engine for science data stream processing. Takes the EarthScope Array Network Facility (ANF) processes and recreates them using SQL technologies and the SQLstream engine. | Aug 09 to Dec 09 | completed | SoW |
SA-Report |
| Unstructured Grid Services | UCSD, OPeNDAP, UW | Integrate Gridfields library and operators for manipulating unstructured gridded data in a data stream based network. Integrate the Hyrax data server backend (C++) with AMQP messaging. | Sep 09 to Dec 09 | completed | SoW |
DM-Report |
| Visualization Workflow Pipeline |
UCSD, UW, Microsoft |
Deploying a Microsoft Trident workflow engine leveraging the Microsoft Azure cloud with workflows feeding ocean model data through a visualization pipeline for UW COVE client. The architecture demonstrates an application architecture that spans client, server, and cloud using the Microsoft Platform. |
completed | SoW |
DM-Report | |
| Instrument Development Kit | OOI IOs | Will make use of the Sensing & Acquisition subsystem technology IaaS (Instrument-as-a-Service) for as instrument agent framework. The actual list of instruments is TBD but likely includes a) Hydrophones b) Semantic Sensors c) Meteorological Station and CDT sensors |
planned |
|
Previous Completed Prototypes
Most of the prototypes prior to FDR (Final Design Review, Nov 2008) were exploratory and conducted as a means of risk mitigation for the risk of integrating heterogeneous leading edge technologies. The attached reports contain findings and lessons learned.
| Prototype Name | Participants | Technical Scope, Comments | Status | Results |
|---|---|---|---|---|
| ESB & Instrument Integration Strategy | OOI/Looking, MBARI, NCSA |
A service and message oriented communication platform based on Mule ESB and ActiveMQ. Integrates Kepler workflows, SIAM and Antelope based instruments, Google Maps web interfaces, data source selection. Executed in a VMware virtual machine environment. Includes a distributed ESB deployment installation providing secure access to streaming instruments (MBARI, NCSA installation). |
completed 2007 |
COI-Prototype Report, COI-Prototype Design |
| Continuous Visualization Pipeline | OOI/Looking, SCCOOS |
A ROMS numerical ocean model (JPL) provides output data products via OpenDAP to a pipeline of VTK (Visual Tool Kit), OpenSceneGraph, OSSIMPlanet, and to Google Earth, using COLLADA. | completed 2008 |
Prototype Report |
| High Availability Strategy Phase 1: Virtual Cluster | OOI | All phases of the high availability strategy prototypes address high availability, local clusters, cloud deployments, in a global network network environment. Phase 1 provides a deployment of multiple instances of a MySQL clustered database system, Confluence and Alfresco applications and the ajproxy load balancer within VMware and XEN virtual machine environments. Other technologies involved include Tangsaw, Spring framework |
completed 2008 |
Prototype Report |
| High Availability Strategy Phase 2: Cloud Deployment | OOI, Globus/Nimbus (ANL), CFT |
Based on phase 1 technologies. Works out the deployment pipeline and how to package and contextualize images. Adds Globus/Nimbus (ANL) for contextualization of instantiated virtual machine images, the CohesiveFT Elastic Server platform for virtual machine image generation from software packages and Amazon EC2 and S3 services as cloud execution environments. |
completed 2008 |
Prototype Report |
| High Availability Strategy Phase 3: Global Traffic Mgt & Resource Load balancing | OOI, Optiputer |
Based on phases 1 and 2. Combines the virtual cluster and cloud deployment prototypes as execution platforms. Combined with Cisco global site selection and load balancing network technologies | in progress (2008) |
Prototype Report |
| Instrument Coupling with Ocean Modeling | OOI/Looking, SCCOOS |
CalPoly mobile sensor network (Remus) and ROMS numerical models |
completed (2008) |
Prototype Report |
| Data Distribution Network Version 1 | OOI, NOAA (IOOS, TBD) |
Initial version of the DDN, released Jan 2009. Combines the technologies and implementations of ERDDAP, THREDDS Data Server (TDS), RabbitMQ. Deployed redundantly and scalably in a cloud execution environment based on Amazon EC2/S3. | completed (2009) |
DDN Design Draft |