Google Compute Engine launched with up to 96 CPU cores and 624 GB of memory
Google Compute Engine has announced an offering with 64 CPU cores and 416 GB of memory. Google has thus doubled the memory Compute Engine previously offered with 32 cores. “These machine types run on Intel Xeon Scalable processors (codenamed Skylake), and offer the most vCPUs of any cloud provider on that chipset. Skylake in turn provides up to 20% faster compute performance, 82% faster HPC performance, and almost 2X the memory bandwidth compared with the previous generation Xeon,” Google said in it announcement. Users can also adjust their workload requirements with custom CPU and memory configurations in case they don’t require that much power. In future, Google is even considering products that deliver up to 4TB of memory.
Edward merge into Tensorflow
Edward, a Python library for probabilistic modeling, inference and criticism, has announced its official merger into TensorFlow. Dustin Tran, who leads the development of Edward, announced that for now Edward will be in the contrib module to avoid redundancy with other submodules. “We’re not sure if all of Edward’s features will be in TensorFlow just yet: for example, it’s unclear where to put Edward’s precise PPL. That said, expect that in this move many new innovations in Edward’s design will appear as we make programmable inference far more flexible, more generally compatible with hardware and distributed choices, and most importantly, more accessible by researchers and applied MLers alike,” Dustin said in the official announcement.
Other Data Science News
PostgreSQL 10 released: logical replication, declarative table partitioning key features
PostgreSQL Global Development Group has announced the release of PostgreSQL 10. The latest version includes several additions that were long anticipated such as native logical replication, declarative table partitioning, and improved query parallelism. “Our developer community focused on building features that would take advantage of modern infrastructure setups for distributing workloads,” said Magnus Hagander, a core team member of the PostgreSQL Global Development Group. The versioning for PostgreSQL has henceforth been revised to “x.y” format, meaning the next minor release will be 10.1 and next major release will be 11.
Microsoft Azure Functions adds support for Java
PyPy v5.9 released with added support for Pandas and NumPy
PyPy has announced the release of its version 5.9, and it now supports Pandas and NumPy too. PyPy 5.8 was released earlier this year in June, where the growing community of PyPy users had reported cases of bugs and other issues. The latest version has several incremental improvements, and the PyPy team has advised that its users go for an update to resolve several ongoing performance issues. According to the announcement, PyPy has released both PyPy3.5 v5.9 (a beta-quality interpreter for Python 3.5 syntax) and PyPy2.7 v5.9 (an interpreter supporting Python 2.7 syntax). NumPy and Pandas now work on PyPy2.7 (together with Cython 0.27.1). CFFI, which has been updated to 1.11.1, now supports complex arguments in API mode, as well as char16_t and char32_t and has improved support for callbacks.