狠狠撸

alluxio storage big data data orchestration open source cloud presto machine learning distributed computing hybrid cloud spark file system cloud computing model training data management summit deep learning caching memory analytics data architecture tachyon project data platform aws infrastructure artificial intelligence cloud storage alluxio day hadoop performance multi cloud hive gpu separation of compute and storage hdfs llm compute ai s3 apache spark distributed systems data loading kubernetes data analytics data engineering sql aws s3 distributed storage gpu utilization data alluxio engineering pytorch distributed data caching meetup emr object store data locality data lake tachyon tech talk object stores release architecture rocksdb tensorflow intel computer fuse posix query engine ray google dataproc unified namespace data lakes cloud bursting nlp trino uber cloud migration data science orchestration local cache cache facebook office hour metadata raft object storage memory centric compute storage separation overview scale hybrid cloud bursting apache ozone nvidia tencent software development database software engineering ml community demo ceph gpu analytics use case software fine tuning apache hudi ecommerce deep learning applications baidu analytics zoo zookeeper scalable poshmark product school microsoft amazon emr python jd memory-centric kyligence nfs rakuten olap structured data management data stores grpc preference tuning ai infra conference data stack fluid alibaba amazon web services data ecosystem data warehouse on-prem zero copy bursting under file system apache iceberg presto caching datasapiens distributed query mesosphere bigdata tutorial qunar unified 2.0 preview global namespace strata multi-tiering walmartlabs data unification virtualization decoupling compute and storage financial services etl datawarehouse distributed system developer developers in-memory storage qiniu elastic mapreduce sogou tachyon nexus business intelligence ctrip momo talking data pingo amplab amazon parquet security storage system analytics and ai cloud architecture twitter virtual file system apache ranger hybrid big data netapp bilibili data tagging open data platform metadata management shadow cache tiktok cache layer prometheus metrics grafana optane persistent memory raptorx disaggregated storage rapids accelerator data lake analytics dask aspect analytics webinar terraform eks model inference inferencing optimization cpu deployment anyscale hpc nvme zoom user experience a/b testing experimentation testing statistics visual search milvus vector database high performance computing framework computing generative ai cv api model traiing devops transparent uri product release t3go high-performance metadata services structured data services catalog service spark workloads remote data software testing unified data zero copy hybrid bursting mapr cloud workloads dc/os object store analytics on-premise compute e-commerce datasets pipeline api usability concurrency iceberg netflix alibaba cloud gene computing structured data search queries ryte walkme unisound atlas starburst robinhood data catalog paypal gimel sql workloads jd.com distributed applications ing tech dataproc google cloud hybrid data lake helixa comcast china unicom aunalytics hub hybrid shannondb storagequery s3 api analytic workloads public cloud high performance zero-copy burst