Chapter 8. The basics of BLOBs
This chapter covers
- How files are currently shared in retro systems
- How Windows Azure allows us to store files (woo hoo, go Azure)
- How to consume the BLOB storage service
In case you didn’t bother reading the blurb at the beginning of part 3, in this chapter (and the next couple of chapters), we’ll be looking at how you can store files in Windows Azure’s highly scalable, fault tolerant, binary-file storage system (otherwise known as the BLOB storage service).
Definition
BLOB stands for binary large object. The term has been stolen from the world of relational databases where it used to describe the storage of binary data (such as an image or an MP3 file) in a single entity. We wish they’d used BinLob as the acronym. It more accurately describes what happens when a DBA discovers you stuck a terabyte of data in a single row of his database.
In this chapter, we’re going to answer the following questions:
- Why is storing files in a typical scaled-out system so hard?
- How does the BLOB storage service address typical scaling issues?
- How does the BLOB storage service work?
- How can you can get your tools out and start developing against it?
- How do you store BLOBs in the production system?
Before you can appreciate the beauty of the BLOB service, you need to get a little insight into how you might solve the problem of storing files that can be accessed by multiple servers in a scalable fashion.