Nvidia to buy Mellanox for $6.9 billion in data centre push

Nvidia Corp. agreed to buy chipmaker Mellanox Technologies Ltd. for $6.9 billion, gaining expertise to help it push into the growing market for data center components.

The $125-a-share cash offer for Mellanox is a 14 % premium to Friday’s close of $109.38. But traders may not be sold on the idea that the deal will go through, with Mellanox trading up 8.6 % to $118.62, still short of the bid from Santa Clara, California-based Nvidia. That gap may reflect concern that chip deals have foundered on regulatory approval delays and blocks amid a trade dispute between China and the U.S.

Nvidia’s biggest-ever acquisition is aimed at accelerating momentum for one of Chief Executive Officer Jensen Huang’s most successful initiatives. The company’s founder built a multi-billion-dollar business in under three years by persuading owners of data centers that his graphics chips are the right solution for processing the increasingly large amounts of information needed for artificial intelligence work, such as image recognition.

“The data center is more important than ever,” Huang said in an interview. “This combination allows us to innovate faster.”

Nvidia is said to have won a bidding process for the American-Israeli company, which makes chips used to speed the flow of information across computer servers, beating out rivals including Intel Corp. Mellanox’s market value, now at about $5.9 billion, started to run up last year when activist investors took stakes and talk that it was up for sale emerged. Shares of the company, which is based in Yokneam, Israel and Sunnyvale, California, have risen 66 % from their October trough and 18 % this year before the deal was announced. Nvidia shares gained 4.8 % at 11:38 a.m. Monday in New York to $157.79.

The acquisition process was ‘‘very competitive,’’ Huang said. Once complete, the combination is expected to be immediately accretive to profit and free cash flow, Nvidia said.

The growing reams of data generated means work on AI and large databases need to be split between multiple computers. Simply using a faster processor isn’t enough, Huang said. To deal with this, data centers in the future will be built as though they are single giant computers “with tens of thousands of compute nodes,” requiring inter-connections that let them work in parallel. Nvidia will use its newly acquired technology to make those giant warehouses full of machinery more efficient and effective, he said.

The deal may signal a resumption of consolidation in the $470 billion semiconductor industry, which has been reshaped over the past five years as companies combined to gain scale while battling rising costs and shrinking customer lists.

“This could reinvigorate M&A discussions across the entire sector. Net: while the transaction would be a positive, given the size of the company, we wouldn’t view it as a transformation,” RBC Capital Markets analyst Mitch Steves wrote in a note to clients. “A deal would be more impactful for the broader semiconductor industry.”