Whereas it could not carry the tag of ‘Silicon Valley of India’ like Bangalore, Chennai, situated simply 300 kilometres away, has emerged as a dominant pressure within the nation’s rising AI ecosystem. With 19 knowledge centres, town ranks second in India for operational knowledge centre capability, behind Mumbai, which has 38 knowledge centres.
As of June final yr, the entire operational knowledge centre capability in Chennai was estimated to be 88 megawatts (MW). Since then, main new institutions from CtrlS (72 MW knowledge centre) and Sify have additional elevated this capability.
The Coastal Benefit for Knowledge Centres
Mumbai and Chennai dominate India’s knowledge centre panorama on account of their coastal geography, which makes them greatest suited to undersea cable touchdown stations. These amenities join the in depth ocean‑spanning fibre-optic strains that carry worldwide web and cellphone site visitors on to terrestrial networks.
On Thursday, on the inauguration of Sify’s Chennai 02 ‘AI-ready’ knowledge centre, which has a capability of over 130 MW, CFO Vijay Kumar stated that the 2 cities perform like “worldwide airports for knowledge”.
“For the web, the 2 [major] gateways within the nation are Mumbai and Chennai, the place all of the subsea cable programs come and land on each side,” stated Kumar, indicating the significance of undersea cable programs in knowledge switch. “Chennai, being nearer to the cable touchdown programs, presents low latency and excessive operational effectiveness,” he informed AIM in an unique interplay.
Main gamers, corresponding to NTT Knowledge, Bharti Airtel, Tata Communications, and Sify, have outfitted their Chennai knowledge centres with on‑web site cable touchdown stations. The Chennai 02 facility’s touchdown station can deal with knowledge transfers at charges of as much as 4 petabytes per second.
There are seven lively undersea cables in Chennai, with three extra underneath development. These hook up with nineteen international locations worldwide, in addition to different cities in India. Connections to main knowledge hubs embrace Europe (France and Italy) and Southeast Asian international locations corresponding to Indonesia, Malaysia, Singapore, and Thailand.
Whereas quite a few knowledge centre initiatives are underway, analysis from JLL estimates that town’s operational capability will double by subsequent yr. Mordor Intelligence estimates that the general capability might exceed 551.15 MW by 2030.
Although India has quite a few coastal cities the place knowledge centres may thrive, Chennai stands out alongside Mumbai for its distinctive development on this sector. This dominance additionally stems from Chennai’s strong enterprise infrastructure, in depth expertise pool, thriving IT ecosystem, and rising variety of world functionality centres (GCCS).
Chennai accounts for 15% of India’s complete IT workforce, and Tamil Nadu ranks third within the nation when it comes to software program exports. Notable Indian IT giants corresponding to Infosys, Wipro, and Tech Mahindra have established places of work within the metropolis, apart from world companies like Accenture, Amazon, Microsoft, IBM, Oracle, and others.
Moreover, the state hosts greater than 250 GCCs. Chennai can also be broadly recognised because the ‘SaaS capital of India’.
Kevin Imboden, world director of market analysis at EdgeConnex, that has arrange an information centre within the metropolis in collaboration with Adani Enterprises, stated, “Whereas different cities throughout India obtained preliminary curiosity as knowledge heart hubs, whether or not for ties to giant headquarters (Mumbai), the IT trade (Bengaluru), or the nationwide authorities (Delhi and the important thing Noida space), Chennai embodies elements of all these cities, with a robust and multifaceted array of industries calling the world house [for data centre establishments].”
Apart from, the state authorities actively helps knowledge centre growth by way of its 2021 coverage. The Tamil Nadu Knowledge Centre Coverage presents varied monetary advantages for firms establishing knowledge centres within the state. These embrace tax concessions, electrical energy subsidies, land value reductions, and extra monetary help to traders.
Furthermore, there’s additionally a booming ecosystem of infrastructure that may deal with AI-related workloads in Chennai. Final yr, Sify was the primary within the nation to realize NVIDIA’s DGX-ready knowledge centre certification for liquid cooling.
This certification validates that knowledge centre parts, together with infrastructure, software program platforms, and managed providers, are totally optimised for NVIDIA DGX AI programs. These turnkey options combine high-performance GPUs, storage, and software program right into a cohesive bundle for AI workloads.
Quickly after, the corporate additionally introduced the Sify Cloudinfinit +AI, a platform that gives GPU-as-a-service for enterprise customers to help varied AI, deep studying and compute-intensive workloads. The corporate supplies providers for India’s main banks, manufacturing hubs, and a number of authorities organisations.
Lately, E2E Cloud unveiled India’s most in depth GPU clusters, which embrace two high-performance clusters, every that includes 1,024 NVIDIA H200 GPUs, positioned in Delhi NCR and Chennai. Moreover, establishments like IIT Madras and the Institute of Mathematical Sciences (IMSc) in Chennai function superior GPU clusters to help AI analysis and workloads.
“Going ahead, any new infrastructure shall be an AI-ready knowledge centre to host excessive workloads,” stated Kumar. “It can be crucial that infrastructure ought to precede the event of functions,” he added, indicating how knowledge facilities are able to host the calls for of upcoming AI workloads and functions.
IIT-M is the X Issue
Apart from {hardware} infrastructure, the presence of the Indian Institute of Expertise Madras (IIT-M) within the metropolis not solely supplies a gifted pool of engineers however has additionally undertaken a number of analysis initiatives and initiatives in AI.
For instance, the Wadhwani Faculty of AI (WSAI) at IITM brings collectively trade leaders, researchers, and college students to collaborate on addressing varied challenges in several areas of synthetic intelligence. Moreover, it presents undergraduate and postgraduate applications in AI and knowledge science.
Most of those analysis initiatives arising from the establishment are centered in the direction of constructing AI for India. In a dialog with AIM, Balaraman Ravindran, head of WSAI, stated the varsity is specializing in creating protocols to check the consistency of language fashions’ reasoning explanations throughout a number of makes an attempt, and creating datasets to judge stereotypical content material particularly in Indian languages slightly than counting on English translations.
Lately, IIT Madras and the IITM Pravartak Applied sciences Basis partnered with California-based Ziroh Labs to ascertain the Centre of AI Analysis (CoAIR) to resolve India’s compute accessibility challenges utilizing AI fashions optimised for CPUs and edge units.
Ziroh Labs launched Kompact AI at IIT Madras – a platform that runs foundational AI fashions on CPUs as an alternative of GPUs. Developed by Indian engineers, it already helps 17 optimised fashions (together with DeepSeek, Qwen, and Llama) and aligns with India’s ‘AI for All’ initiative.
Moreover, AI4Bharat, the AI lab at IIT Madras, just lately launched IndicTrans3-beta, a state-of-the-art (SOTA) multilingual translation mannequin designed to help translations throughout 22 Indic languages.
Final yr, the AI lab additionally launched BhasaAnuvaad, a speech translation dataset tailor-made for Indian languages, boasting protection throughout 13 languages and roughly 44,400 hours of audio. This was the most important publicly accessible speech translation useful resource of its type for Indian linguistic range.
All issues thought of, Chennai’s analysis and growth initiatives, together with highly effective AI-ready infrastructure hubs, are set to play a key function within the IndiaAI mission, which focuses on constructing a frontier AI mannequin within the nation.
The put up India’s AI Future Will Trip on The Chennai Categorical appeared first on Analytics India Journal.
