델타: 체인 복제 기반 고가용성·강한 일관성 스토리지

By Kumar Mrinal , Binbin Lu Over the years, Meta has invested in a number of storage service offerings that cater to different use cases and workload characteristics. Along the way, we've aimed to reduce and converge the systems in the storage space. At the same time, having a dedicated solution for critical package workload makes everyone happier. Having this in place is necessary for our disaster recovery and bootstrap strategy. This realization, coupled with a business need to provide storage for Meta’s build and distribution artifacts, led to the inception of a new object storage service — Delta. Consider Delta’s positioning in the Meta infrastructure stack (below). It belongs at the very bottom, providing the basic primitive required for the availability and recoverability of the rest of the infrastructure. For bootstrap systems, complexity should be introduced only if it makes the solution more reliable. We’re only minimally concerned with the performance and efficiency of the solution. Another consideration for bootstrap systems involves the bootstrap itself. This process, by which engineers can access a small set of machines and restore the rest of our infrastructure, helps us get the product back up and working for people using it. Lastly, the bootstrap data needs to be backed up for recovery in case disaster strikes. In this post, we will discuss the goals for Delta, the main concepts that govern Delta’s architecture, Delta’s production use cases, its evolution as a recovery provider, and future work items. What is Delta? Delta is a simple, reliable, scalable, low-dependency object storage system. It features only four high-level operations: put, get, delete, and list. Delta trades latency and storage efficiency in favor of simplicity and reliability. Since it’s horizontally scalable, Delta takes on minimal dependencies with appropriate failover strategies for soft dependencies in place. Delta is not a : General purpose storage system: Delta’s core tenets are resiliency and data protection. It’s designed specifically to be used by low-dependency systems. Filesystem: Delta acts as a simple object storage system. It doesn’t intend to expose filesystem semantics like Posix etc. System optimized for maximum storage efficiency: With resiliency as its primary tenet and focus on critical systems, Delta doesn’t intend to optimize for storage efficiency, latency, or throughput. Delta’s architecture Delta has productionized chain replication , an approach to coordinating clusters of fail-stop storage servers. It intends to support large-scale storage services that exhibit high throughput and availability, without sacrificing strong consistency guarantees. Before diving deeper into how Delta leverages chain replication to replicate client data, let’s first explore the basics of chain replication. Chain replication Fundamentally, chain replication organizes servers in a chain in a linear fashion. Much like a linked list, each chain involves a set of hosts that redundantly store replicas of objects. Each chain contains a sequence of servers. We call the first server the head and the last one the tail. The figure below shows an example of a chain with four servers. Each write request gets directed to the head server. The update pipelines from the head server to the tail server through the chain. Once all the servers have persisted the update, the tail responds to the write request. Read requests are directed only to tail servers. What a client can read from the tail of the chain replicates across all servers belonging to the chain, guaranteeing strong consistency. Chain replication vs. quorum replication Now that we have provided an overview of what chain replication entails, let’s explore how chain replication fares against other widely used replication strategies. Storage efficiency: Chain replication clearly does not offer the most storage-efficient replication strategy. We store redundant copies of the whole data set on all hosts in a chain. A comparatively efficient approach would involve intelligently replicating fragments of data using erasure coding techniques. Fault tolerance: In an optimal bucket layout, chain replication can provide similar or better fault tolerance than quorum-based replication mechanisms. Why? A chain with `n` nodes can tolerate failures up to `n - 2` nodes without compromising on availability. On the contrary, for quorum-based replication systems, at least `w` hosts must be available to serve writes. Additionally, `r` hosts must be available to serve reads. Here `w` and `r` represent the write quorum size and read quorum size, respectively. Performance: In replication strategies (like primary backup), all backup servers can serve reads. This increases read throughput. In the native idea of chain replication, only the chain tail can serve reads. (We optimized this bit and will share details in the apportioned queries section later in this post.) Much like quorum-based replication mechanisms, in chain replication all writes are directed to the primary (the head of the chain). But in chain replication, writes are only responded to after all hosts in the chain have acknowledged the update. Thus, chain replication has higher write latency on average in comparison with quorum-based replication mechanisms. Quorum consensus: Quorum-based systems need complex consensus and leader election mechanisms to maintain quorum in the system. In contrast, the scope of quorum consensus in a chain-replication based system gets narrowed down to the much simpler, chain-host mapping. For example, the chain head always serves as a leader for processing writes, without the need for an explicit leader election. Considering the above differences, chain replication clearly fails to offer the most storage-efficient way to replicate data across machines. Additionally, it yields higher average write latency in comparison to quorum-based systems, as we consider writes successful only when all links in a chain have persisted the update. However, it’s very simple while offering similar fault tolerance and consistency guarantees. The anatomy of a Delta bucket Now that we have a preliminary understanding of chain replication, let’s talk about how Delta leverages it to replicate data across multiple servers. Each Delta bucket above includes several chains. Each chain usually consists of four or more servers, which can vary based on the desired replication factor. Each chain itself acts as a replica set and serves a slice of data and traffic. It can be thought of as a logical shard of a client data set. Servers in a particular chain get spread across different failure domains (power, network, etc.). Doing so guarantees durability and availability of client data if servers in one or more failure domains remain unavailable. We maintain a bucket config, the authoritative chain-host mapping for the layout of the bucket. When we add or remove servers and chains from the bucket, the bucket config gets appropriately updated. When clients access an object within a Delta bucket, a consistent hash of the object name selects the appropriate chain. Writes are always directed to the head of the appropriate chain. It writes the data to the local storage and forwards the write to the next host in the chain. The write is acknowledged only after the last host in the chain has durably stored the data on local media. Reads are always directed to the tail of the appropriate chain. This guarantees that only fully replicated data is visible and readable, thereby guaranteeing strong consistency. Delta supports horizontal scalability by adding new servers into the bucket and smartly rebalancing chains to the newly added servers without affecting the service’s availability and throughput. As an example, one tactic is to have servers with the most chains transfer some chains to new servers as a way to rebalance the load. New bucket layouts would still follow the desirable failure-domai