Today in Cloud

I covered on the Structure blog today Yahoo’s decision to open source its S4 project, which the company calls “real-time MapReduce,” ostensibly because it relies upon Hadoop clusters for processing. I would argue this is actually quite a big deal considering the focus that vendors like IBM, SAP, Teradata and Oracle have been placing on real-time processing lately. So, here’s my question: If someone were to incorporate S4 into its Hadoop distribution or development platform, would it have a meaningful impact on the types of jobs organizations are thinking about when contemplating Hadoop?