Jezper - Fotolia
Broadcom Brocade deal leaves questions
This week, bloggers weigh the possible outcomes of the Broadcom Brocade deal and discuss cybersecurity lessons fueled by the 2016 U.S. election.
Drew Conry-Murray, writing in Packet Pushers, sees open questions in the wake of the Broadcom Brocade acquisition. Broadcom purchased Brocade Communications Systems Inc. last month and quickly announced it would spin off Brocade's IP networking business -- spanning routers, switches and Ruckus Wireless, the WLAN vendor Brocade purchased earlier this year. Looking at the IP components of Brocade's portfolio, Conry-Murray reports that the unit earned Brocade $601 million in 2015 compared to $1.2 billion the vendor earned from its Fibre Channel business. The purchase of Ruckus, meantime, cost Brocade $1.5 billion, yet the wireless business generated $73 million in revenue in Q3, Conry-Murray said.
In Conry-Murray's view, Broadcom Brocade is likely to pursue one of four different options to spin off the IP unit. A private equity firm may buy the IP networking business or a different networking vendor may step in as purchaser. Alternatively, the networking business may be broken up piecemeal or spun off. "If I were a betting man, I'd put my money on private equity. With interest rates so low, raising funds is cheap and there's a lot of cash out there looking for returns," Conry-Murray said.
Read more of Conry-Murray's thoughts on Broadcom Brocade.
Election lessons for cybersecurity
Jon Oltsik, an analyst at Enterprise Strategy Group in Milford, Mass., says that he -- like millions of others -- believed Hillary Clinton was the most likely to win the U.S. presidency. Data and polling supported that notion, but Donald Trump's ultimate victory reflected illuminated vast shortcomings in how that polling data was collected. In essence, powerful algorithms failed because of poor data gathering. Oltsik fears that cybersecurity tools that use the same type of algorithms, artificial intelligence and data gathering to underpin their services are equally suspect. "I'm afraid that cybersecurity data models may suffer the same fate because there simply aren't enough experienced cybersecurity professionals available with situational awareness and data expertise to make all of these models robust," he said.
According to ESG research, 46% of organizations say they have a problematic shortage of cybersecurity skills while a separate survey indicated that 56% of cybersecurity professionals believe the level of training they receive from their employers is inadequate. "New types of cybersecurity analytics depend upon two types of people: data scientists who can build the models and cybersecurity subject matter experts who can feed the models with the right assumptions, data, and situational awareness," Oltsik commented. "Unfortunately, the ESG/ISSA data demonstrates that there just aren't enough of these latter folks to go around," he added.
Explore more of Oltsik's thoughts on cybersecurity.
How reliable are clustered systems?
Ivan Pepelnjak, blogging in ipSpace, looked back on a conclusion he had previously shared on his website. "I haven't seen any hard data, but intuition suggests that apart from hardware failures a standalone firewall might be more stable than a state-sharing firewall cluster," he said at the time. Guillaume Sachot, an employee at a web hosting company responded to Pepelnjak's idea, saying that he had seen high-availability appliances fail more often than clustered ones.
Sachot said that he had seen this pattern of failure for almost every type of high-availability program, such as DRBD instances, Pacemaker and shared file systems. "It fails, fails and fails. There are bugs, but also configuration issues/bad sync, network flapping that breaks replication and puts all instances in standby/secondary mode. ... In the end, a non-redundant VM on a nonclustered host seems to be the service that lasts the longest." Sachot recommended non-redundant VMs running on a cluster available to be manually triggered or to automatically cover host-failure. Furthermore, for people in search of high-availability, Sachot suggested using multiple locations or independent providers.
Explore more of Pepelnjak and Sachot's thoughts on reliability.
