Situational Awareness Response and Help AI
Computing with SARAHAI
SARAHAI is a combined video analytics and video management and networking platform. Large numbers of internet protocol (IP) cameras exist for security and other applications that are siloed by disparate and proprietary systems and unable to share or manage content across systems. Video analytics for many feeds is a compute-intensive job, where the performance goal is high rate of job completion rather than low turnaround time of individual jobs. SARAHAI includes schema for local distributed data storage. SARAHAI has a federated edge computing architecture but can also support cloud implementation across IaaS.
SARAHAAI can be used in two ways, depending on where the worker nodes come from:
In a local area network onsite or across the Internet in a wide area network configuration with IPSEC and VPN encrypted traffic.
Within a hosted data center/cloud utilizing IaaS (Infrastructure as a Service).
SARAHAI can run all existing HTC applications, including those that use GPUs and/or multiple CPU cores. It can use virtual machines to run existing Linux applications and on Windows WSL nodes.
SARAHAI provides mechanisms for job submission and control, designed for performance at scale. However, it can also be used as a back end for existing job-submission systems.
SARAHAI was created to provide data scientists with big computing power at a small cost. Suppose you need, say, 100 TeraFLOPS for 1 year. Here are some ways you can get it:
Use Amazon's Elastic Computing Cloud: $175 Million
Based on $0.10 per node/hour.
Build a cluster: $12.4 Million
This includes power and air-conditioning infrastructure, network hardware, computing hardware, storage, electricity, and sysadmin personnel.
Use SARAHAI: <$125,000
Based on the average throughput and budget of the 6 largest projects.
It takes (very roughly) three man-months to deliver a SARAHAI project: one month of an experienced sys admin, one month of a programmer, and one month of a web developer. Once the project is running, budget a 50% FTE (mostly system admin) to maintain it. In terms of hardware, you'll need a mid-range computing and a fast connection to the commercial Internet.