Download the authoritative guide: Enterprise Data Storage 2018: Optimizing Your Storage Infrastructure
Object storage is a data storage architecture that manages data as objects, as opposed to the traditional way of managing storage as files or blocks. An object has data and metadata, which is customizable. This means it can include plenty of descriptive information, which makes it useful for data analytics.
Enterprises have adopted this storage technology, seeking the major benefits of object storage, which are:
* Scalability: It's easy to scale object storage by adding more storage to the pool. This makes it a good fit for the ever-growing unstructured data such as movies, images, archive, backup.
* Access: The protocol is not sensitive to latency, which makes it good for storing data in multiple regions.https://o1.qnsr.com/log/p.gif?;n=203;c=204650394;s=9477;x=7936;f=201801171506010;u=j;z=TIMESTAMP;a=20392931;e=i
* Agility: Object storage is less complex and requires fewer administrative resources compared to other data storage architectures.
Regulatory archiving in areas like financial, healthcare and legal is one big area of object storage deployment. Another usage group utilizes object access or object architecture-based solutions as a home for large amounts of data that may not be as active as primary high-performance storage such as for database, email and others, including large-scale web files, high-performance compute, video, media and entertainment storage.
IBM acquired Cleversafe Dispersed Storage Network (dsNet) and added its own technology elements to create IBM Cloud Object Storage. As such, it has put together a feature rich object storage solution.
IBM Cloud Object Storage stores large volumes of data while ensuring security, availability and reliability. By using Information Dispersal Algorithms (IDAs) to separate data in unrecognizable “slices” that are distributed across data centers, the transmission and storage of data is made secure. No complete copy of the data resides in any single storage node, and only a subset of nodes needs to be available to retrieve the data on the network. There are three resiliency options: cross region, regional and single data center.
All objects stored are encrypted by default. Data at rest is encrypted with Advanced Encryption Standard (AES) 256-bit encryption and Secure Hash Algorithm (SHA)-256 hash. Data in motion is secured using carrier grade TLS/SSL. IBM Cloud Identity & Access Management (IAM) provides policy control on who can access data in Cloud Object Storage buckets. You can set bucket-level access policies, selectively grant permissions, assign user roles, and control the actions that users and applications can perform.
Services are available in several classes. Standard is for active workloads when data is accessed multiple times a month. Vault is for less-frequently accessed data (once a month or less). Cold Vault is for long-term retention of data that is rarely or never accesses. Flex is for unpredictable data usage patterns. It offers a cap on the combined capacity and retrieval charge.
An archive capability is priced at $0.002 GB/month for data storage and $0.02 GB for data restore, with a minimum duration charge of 180 days. Finally, Aspera high-speed transfer technology is integrated with Cloud Object Storage as a very rapid way to move data into Cloud Object Storage.
Key Use Case: Stores large volumes of data while ensuring security, availability and reliability. Services are available in several classes, depending on how frequently the data is accessed.
Tarmin GridBank Enterprise optimizes data volumes and eliminates data silos by consolidating volumes into a single view while applying storage reduction technologies It utilizes a scalable grid-based architecture to store and manage data at multi-petabyte levels. This makes it possible to provide high performance random access referential metadata in-cache lookups for unstructured data.
GridBank leverages enterprise object storage and adds data management to meet compliance, data protection and other risk management needs. It automatically applies distributed object deduplication and compression to data storage volumes.
Users tend to be data-intensive industries with rapidly growing unstructured data problems that impact data governance and compliance requirements. These industries are also interested in scale out, storage agnostic solutions that can provide a path to the cloud and the ability to leverage analytics. This includes financial services, healthcare, education, oil & gas, life sciences, as well as cloud and managed service providers.
GridBank supports REST API for data management, as well as S3, CDMI and OpenStack compatible APIs for cloud provisioning and integration with existing applications. Data can also be accessed using standard network protocols such as SMB (CIFS), NFS, as well as web access using HTTPS and FTP.
Tarmin GridBank goes beyond the basic technology components found in many common object storage products to create a single integrated data--centric management architecture. GridBank provides direct integration into Exchange, SharePoint and many social media applications, removing another layer of complexity for organizations considering the adoption of a solution that includes object storage.
Key Use Case: Geared for data-intensive industries with rapidly growing unstructured data problems that impact data governance and compliance requirements. Offers a single integrated data--centric management architecture.
Dell EMC Elastic Cloud Storage (ECS) is characterized as software-defined, globally distributed object storage with low cost of ownership. Its use case is mainly as a content repository for cloud-native applications, the Internet of Things, storage as a service, hybrid cloud, consolidated backup and archive deployments.
It is available as a fully configured ECS appliance, as software or as an ECS Dedicated Cloud Service. The appliance is said to be a multi-purpose, software-defined cloud storage platform. The ECS Appliance U Series has configurations that start at 400 TB and can scale very high. The D-Series is all about density: It can provide up to 7.8 PB in a single rack.
Key Use Case: Designed as content repository for cloud-native applications, IoT, storage as a service, hybrid cloud, consolidated backup and archive deployments.
Quantum’s Lattus is aimed at large-scale, long-term archives. The company’s erasure code technology combines data protection with data storage to ensure a single instance of data is protected. Anything from satellite images, CAD drawings, film industry dailies, office documents, DNA sequence data, or reports can be stored in Lattus disk archives. It can scale to hundreds of Petabytes with a flat object namespace.
Data dispersion algorithms can be tuned to spread the data across multiple sites to improve accessibility and availability. This is said to eliminate the need to replicate data. Options include a NAS gateway and Lattus S3 HTTP REST interface. It also integrates with third party applications with StorNext Storage Manager. Similarly, Rocket Arkivio offers intelligent data movement solution for those wishing to leverage Lattus as an active archive tier of storage.
Users can add new storage nodes to the system, and old ones retired, and the data will automatically respread to include the new nodes, and exclude the old nodes. Lattus is best suited for large archives with high or unpredictable growth where business and production users need fast access to the wealth of assets in their archive.
Key Use Case: Combines data protection with data storage to ensure a single instance of data is protected. Integrates with third party applications. Well suited for large archives with high or unpredictable growth.
NooBaa, an object storage startup, is a software-defined data platform for distributed hybrid environments. It allows enterprises to span data either on-premises, on the cloud or in a hybrid and multi-cloud architecture dynamically. Users have full control of redundancy levels, location, and access and they set up different policies per application. Data is accessed via an AWS S3 compatible API. The data itself is deduplicated and inline and compressed for better economy.
Data is also encrypted with encryption keys separated from the data security in the cloud. Data can be spanned, so no single location/cloud has a full copy of the data. It's easy to start with NooBaa, as it is offered in a freemium model for the first 20 TB and the implementation is done by the customer in under five minutes using existing infrastructure or the cloud for risk-free and agile evaluation.
NooBaa can be introduced to existing object and cloud repositories without having to migrate them before using them. The company claims that combining these abilities reduces data silos while improving business flexibility and lowering TCO.
Key Use Case: Enables companies to span data either on-premises, on the cloud or in a hybrid and multi-cloud architecture dynamically. Can be deployed to existing object and cloud repositories without having to migrate them before use.
Plenty of Options for Object Storage
Of course, this overview of object storage providers offers only a portion of available options. Don’t forget about the likes of NetApp StorageGrid, Hitachi Content Platform and Caringo, among others. With so many vendors involved and jockeying for market supremacy, look for plenty of innovation in this field in the years ahead.
But in any case, experts appear clear on the topic: Object storage will continue to evolve similar to how file/NAS accessed storage continues to evolve and get adopted. Which suggests a bright future for object storage going forward.