Future of Cloud Storage AB Periasamy | CTO Gluster, Inc. Thu, June 9, 2011
Storage Transforming to Reflect Compute Multi-tenant / CommoditizedVirtualized Automated shared Standardized Scale on Free Software In the Cloud Scale Out Demand / OpenSource Storage must support the public and private cloud environment • Storage is the Achilles heel of full data center virtualization • Storage performance, availability, capacity, & interface is the Achilles heel of public cloud • Big Data, data migration, multi-site consistency are the Achilles heel of hybrid cloud Storage will look like the computing environment • Storage should be a commoditized, virtualized, centrally managed pool • Monolithic proprietary systems now challenged by nimble and shared network storage • Storage becomes a software problem • The “Google model” of storage Petascale Cloud Filesystem 2
Disk Systems & Storage Interconnects FC (FCP) Infiniband (SRP, iSer, NFSoRDMA) Ethernet (iSCSI, FCoE, AoE, NBD/DRBD, NFS, HTTP) DAS vs JBOD vs SAN SATA vs SAS vs SSD Petascale Cloud Filesystem 3
Gluster Architecture Advantages Software only No metadata server‘Google Storage’ for • Fully distributed architecture, no bottleneckEveryone • Gluster Elastic Hash• Intelligence in the SW High performance global namespace• Leverage commodity HW • Scale out with linear performance • Hundreds of petabytes• Scale-out elastically • 1 GbE, 10GbE• Replication for reliability High availability• Software enables • Replication to survive hardware failure virtualization • Self-healing • Data stored in NFS-like native format Stackable userspace design • No kernel dependencies, simple install • Match specific workload profiles • Early maturity and rich functionality Petascale Cloud Filesystem 9
Evolution of GlusterFS 2006-2009 GlusterFS v1.0 – v3.0 Distributed Filesystem capabilities with self-healing, synchronous replication, stripe, distribute (global name space) 2010 GlusterFS v3.1 Elastic Cloud capabilities 2011 Q2 GlusterFS v3.2 GeoGraphic replication, Enhanced monitoring, Directory level quotas (also works as cloud usage billing APIs) 2011 Q3/Q4 Hadoop HDFS drop-in replacement, Unified File and Object Storage (Amazon S3 compatible) and Near CDP. Petascale Cloud Filesystem 10
Story of Gluster 1st meeting room 1st Office US 1st Office Bengalooru 1st Office Bengalooru Petascale Cloud Filesystem 11
1000s of Community Deployments Petascale Cloud Filesystem 12
Fast Growing Commercial Deployments Petascale Cloud Filesystem 13
GlusterFS & OpenStack VM Image Storage – Answer to VMWare VMFS Unified File & Object Storage – Application Data GeoReplication – Enable Hybrid Clouds Petascale Cloud Filesystem 16
Partners HealthcarePrivate Cloud: Centralized Storage as a Service Problem • Capacity growth from 144TB to 1+PB • Multiple distributed users/departments • Multi OS access - Windows, Linux and Unix Solution • Over 500 TB • GlusterFS Cluster • 9 Sun “Thumper” • Solaris/ZFS/x4500 w/ InfiniBand systems in cluster • Native CIFS/ NFS access Benefits • Capacity on demand / pay as you grow • Centralized management • Higher reliability • OPEX decreased by 10X Petascale Cloud Filesystem 17
Pandora Internet Radio Problem • Explosive user & title growth • As many as 12 file formats for each song • ‘Hot’ content and long tail Solution • Three data centers, each with a six-node GlusterFS cluster• 1.2 PB of audio served • Replication for high availability per week • 250+ TB total capacity• 13 million files Benefits• Over 50 GB/sec peak • Easily scale capacity traffic • Centralized management; one administrator to manage day-to-day operations • No changes to application • Higher reliability Petascale Cloud Filesystem 18
Cincinnati Bell Technology Solutions Problem • Host a dedicated enterprise cloud solution • Large scale VMware environment • Need high availability Solution• Large scale VM • Gluster for VM storage, NFS to clients storage • SAS drives on back-end• Low cost service • Replication for high availability delivery for enterprise customer Benefits• Drastic reduction in • Storage provisioning from 6 wks. to 15 min. provisioning time • Vendor agnostic storage • Low cost of service delivery • Elastic growth Petascale Cloud Filesystem 19
Envoy MediaPublic Cloud: Media Serving on AWS Problem • Limited scalability • Slow response to demand spikes • Manual data management Solution • Four EBS volumes under Gluster global• Targeted media namespace serving • Replication for high availability• 100% AWS hosted • EC2 for compute; S3 for backup• Unpredictable traffic Benefits • No change to application • Content immediately available to all servers • Automatic resource allocation • Lower cost (vs. colo and proprietary options) Petascale Cloud Filesystem 20
Gostou de algum slide específico?
Recortar slides é uma maneira fácil de colecionar informações para acessar mais tarde.