All organizations are looking to the cloud for their compute and data storage needs. Amazon AWS is, by far, the current market leader in terms of revenue as well as number of customers. And its service ecosystem is the most complete. One of the most successful services in its product portfolio is Simple Storage Service (S3). The first service launched by Amazon AWS in 2006, S3 is an object store and, as such, has durability, availability, and cost that is better than any other form of storage offered by this provider.
Glacier, Amazon’s archival object storage service, has an even better price point but is intended for cold archiving only. S3 is also the name of the Application Program Interface (API) for accessing data in this system, and is widely considered the de facto industry standard for object stores.
Object storage is perfect for many use cases and applications, both in the cloud and on premises, and is becoming a very popular target for backup, archiving, content management, file-based applications, big data lakes, and so on. In other words, a large part of applications that deal with unstructured data can easily take advantage of an object store, and this is why it is becoming so popular.
As already mentioned, Amazon S3 is relatively inexpensive compared to other storage options available from Amazon AWS but, on the flipside, its performance is not always consistent. Even more so, the real cost of the service could become an issue for some customers. In fact, the S3 pricing model is quite complex and depends on several factors:
- Type of data protection
- Data locality
- Storage tier
- IO operations
- Data transferred out of AWS (egress)
At the end of the day, it is quite complicated to estimate the cost and it can be difficult to predict how it will evolve over time. Egress costs of public cloud services are what worry CFOs and project managers the most, limiting flexibility in the execution of a multi-cloud strategy.
Although Amazon AWS has a very compelling solution ecosystem, there are niche players with innovative services for vertical markets and use cases that offer interesting alternatives. With this type of multi-cloud approach, the egress cost can become very important and undermine overall infrastructure TCO. As soon as this kind of issue surfaces some end users prefer to repatriate their data on private object stores and get very predictable costs, while others stick with public cloud and its OPEX model. In the latter case, thanks to the success of Amazon AWS S3, many competitors have begun providing similar services while trying to differentiate themselves on price, performance, and functionality.
Not all object stores are born equal though. In this report we will briefly analyze the most interesting services available in North America and Europe focusing on features like:
- Data protection, durability, and availability
- Security aspects
- Geographic availability of the service
- Pricing and business model
- Supported APIs
- Partner ecosystem
- Multi tiering capabilities
Key findings in this report include:
- Different aspects to evaluate when choosing an alternative to Amazon AWS S3, which include performance as well as the business model of market competitors
- Applications and business needs that drive this type of choice; including technical aspects like latency, regional availability of the service, and vicinity to major cloud providers
- Regulations and compliance that could come into play in some industries or countries, affecting the choice of the service provider.
- Market Framework
- Considerations for Using S3 Alternatives
- Vendors to Watch
- Near Term Outlook
- Key Takeaways
- About Enrico Signoretti
- About GigaOm