https://www.nextplatform.com/2017/11/21/marvell-cavium-forge-datacenter-platform/ In-depth coverage of high-end computing at large enterprises, supercomputing centers, hyperscale data centers, and public clouds. Mon, 23 Apr 2018 11:22:17 +0000 hourly 1 https://wordpress.org/?v=6.5.5 https://www.nextplatform.com/2017/11/21/marvell-cavium-forge-datacenter-platform/#comment-95314 Sun, 26 Nov 2017 23:29:16 +0000 https://www.nextplatform.com/?p=35562#comment-95314 Both these companies are past their prime. Cavium has not made any profit for many years and Marvell has lost its relevance for long time. The merger reminds me of two drunks trying to walk together arm in arm. The datacenter market is shrinking and hyperscale is the wave of the future. I don’t see any way these two could dislodge Intel in Hyperscale market. All the ARM chip makers never mention that ARM is awful compared to x86 in floating point performance and even the best ARM cores can come only close to Intel in integer performance. They can never beat Intel. Cavium got Vulcan project (ARMv8) team from Broadcom when Broadcom dumped them but it has not yielded any result except some announcements in HPC world. In HPC world, only large system integrators and software developers make money, not the chip makers. Broadcom Vulcan was a failed project that could never deliver anything on time. So there is no way Broadcom is going to regret anything by unloading Vulcan. So this merger is clearly aimed at consolidation and staying alive a bit longer. It won’t change the future .

]]>