Saurabh Kulkarni

Kapil Vaswani, Stavros Volos, Cédric Fournet et al.

We present IPU Trusted Extensions (ITX), a set of experimental hardware extensions that enable trusted execution environments in Graphcore's AI accelerators. ITX enables the execution of AI workloads with strong confidentiality and integrity guarantees at low performance overheads. ITX isolates workloads from untrusted hosts, and ensures their data and models remain encrypted at all times except within the IPU. ITX includes a hardware root-of-trust that provides attestation capabilities and orchestrates trusted execution, and on-chip programmable cryptographic engines for authenticated encryption of code and data at PCIe bandwidth. We also present software for ITX in the form of compiler and runtime extensions that support multi-party training without requiring a CPU-based TEE. Experimental support for ITX is included in Graphcore's GC200 IPU taped out at TSMC's 7nm technology node. Its evaluation on a development board using standard DNN training workloads suggests that ITX adds less than 5% performance overhead, and delivers up to 17x better performance compared to CPU-based confidential computing systems relying on AMD SEV-SNP.

Ajay Kulkarni, Saurabh Kulkarni

Open source is the future of technology. This community is growing by the day; developing and improving existing frameworks and software for free. Open source replacements are coming up for almost all proprietary software nowadays. This paper proposes an open source application which could replace Skype, a popular VoIP soft phone. The performance features of the developed software is analyzed and compared with Skype so that we can conclude that it can be an efficient replacement. This application is developed in pure Java using various APIs and package and boasts features like voice calling, chatting, file sharing etc. The target audience for this software will initially only be organizations (for internal communication) and later will be released on a larger scale.

Chaitanya Rele, Aditya Rathod, Kaustubh Natu et al.

The North Indian Ocean, including the Arabian Sea and the Bay of Bengal, represents a vital source of livelihood for coastal communities, yet fishermen often face uncertainty in locating productive fishing grounds. To address this challenge, we present an AI-assisted framework for predicting Potential Fishing Zones (PFZs) using oceanographic parameters such as sea surface temperature and chlorophyll concentration. The approach is designed to enhance the accuracy of PFZ identification and provide region-specific insights for sustainable fishing practices. Preliminary results indicate that the framework can support fishermen by reducing search time, lowering fuel consumption, and promoting efficient resource utilization.

Ajay Kulkarni, Saurabh Kulkarni, Ketki Haridas et al.

Securing multimedia data has become of utmost importance especially in the applications related to military purposes. With the rise in development in computer and internet technology, multimedia data has become the most convenient method for military training. An innovative encryption algorithm for videos compressed using H.264 was proposed to safely exchange highly confidential videos. To maintain a balance between security and computational time, the proposed algorithm shuffles the video frames along with the audio, and then AES is used to selectively encrypt the sensitive video codewords. Using this approach unauthorized viewing of the video file can be prevented and hence this algorithm provides a high level of security. A comparative study of the proposed algorithm with other existing algorithms has been put forward in this paper to prove the effectiveness of the proposed algorithm.