I am having hard time to grab what etcd (in CoreOS) really does, because all those "distributed key-value storage" thingy seems intangible to me. Further reading into etcd, it delves into into Raft consensus algorithm, and then it becomes really confusing to understand.
Let's say that what happen if a cluster system doesn't have etcd?
Thanks for your time and effort!
As someone with no CoreOS experience building a distributed system using etcd, I think I can shed some light on this.
The idea with etcd is to give some very basic primitives that are applicable for building a wide variety of distributed systems. The reason for this is that distributed systems are fundamentally hard. Most programmers don't really grok the difficulties simply because there are orders of magnitude more opportunity to learn about single-system programs; this has really only started to shift in the last 5 years since cloud computing made distributed systems cheap to build and experiment with. Even so, there's a lot to learn.
One of the biggest problems in distributed systems is consensus. In other words, guaranteeing that all nodes in a system agree on a particular value. Now, if hardware and networks were 100% reliable then it would be easy, but of course that is impossible. Designing an algorithm to provide some meaningful guarantees around consensus is a very difficult problem, and one that a lot of smart people have put a lot of time into. Paxos was the previous state of the art algorithm, but was very difficult to understand. Raft is an attempt to provide similar guarantees but be much more approachable to the average programmer. However, even so, as you have discovered, it is non-trivial to understand it's operational details and applications.
In terms of what etcd is specifically used for in CoreOS I can't tell you. But what I can say with certainty is that any data which needs to be shared and agreed upon by all machines in a cluster should be stored in etcd. Conversely, anything that a node (or subset of nodes) can handle on its own should emphatically not be stored in etcd (because it incurs the overhead of communicating and storing it on all nodes).
With etcd it's possible to have a large number of identical machines automatically coordinate, elect a leader, and guarantee an identical history of data in its key-value store such that:
No etcd node will ever return data which is not agreed upon by the majority of nodes.
For cluster size x any number of machines > x/2 can continue operating and accepting writes even if the others die or lose connectivity.
For any machines losing connectivity (eg. due to a netsplit), they are guaranteed to continue to return correct historical data even though they will fail to write.
The key-value store itself is quite simple and nothing particularly interesting, but these properties allow one to construct distributed systems that resist individual component failure and can provide reasonable guarantees of correctness.
etcd is a reliable system for cluster-wide coordination and state management. It is built on top of Raft.
Raft gives etcd a total ordering of events across a system of distributed etcd nodes. This has many advantages and disadvantages:
Advantages include:
any node may be treated like a master
minimal downtime (a client can try another node if one isn't responding)
avoids split-braining
a reliable way to build distributed locks for cluster-wide coordination
users of etcd can build distributed systems without ad-hoc, buggy, homegrown solutions
For example: You would use etcd to coordinate an automated election of a new Postgres master so that there remains only one master in the cluster.
Disadvantages include:
for safety reasons, it requires a majority of the cluster to commit writes - usually to disk - before replying to a client
requires more network chatter than a single master system
Related
I got a question regarding reinforcement learning. let's say we have a robot that is able to adapt to changing environments. Similar to this paper 1. When there is a change in the environment[light dimming], the robot's performance drops and it needs to explore its new environment by collecting data and running the Q-algorithm again to update its policy to be able to "adapt". The collection of new data and updating of the policy takes about 4/5hrs. I was wondering if I have an army of these robots in the same room, undergoing the same environmental changes, can the data collection be sped up so that the policy can be updated more quickly? so that the policy can be updated in under 1 hour or so, allowing the performance of the robots to increase?
I believe you are talking about scaling learning horizontally as in training multiple agents in parallel.
A3C is one algorithm that does this by training multiple agents in parallel and independently of each other. Each agent has its own environment which allows it to gain a different experience than the rest of the agents, ultimately increasing the breadth of your agents collective experience. Eventually each agent updates a shared network asynchronously and you use this network to drive your main agent.
You mentioned that you wanted to use the same environment for all parallel agents. I can think of this in two ways:
If you are talking about a shared environment among agents, then this could possibly speed things up however you are likely not going to gain much in terms of performance. You are also very likely to face issues in terms of episode completion - if multiple agents are taking steps simultaneously then your transitions will be a mess to say the least. The complexity cost is high and the benefit is negligible.
If you are talking about cloning the same environment for each agent then you end up both gaining speed and a broader experience which translates to performance. This is probably the sane thing to do.
"Inability of MySQL to scale write requests beyond one node became a killer problem as data volumes grew leaps and bounds. MySQL’s monolithic architecture essentially forces application-level sharding.The development and operational complexity grows exponentially when the number of such instances grow from 1 to 100s and thereafter explode into 1000s."
So, If I use sharding for scaling up the application then
I cannot use the database itself for any cross-shard JOINs and transactions. It seems true but it Is it True ??
What's the better and efficient way to scaling the application as well as the database for say 10k concurrent requests and more. ??
I think going for microservices is not the solution here or is it ??
The answer is more nuanced than simply true or false. A horizontal
sharding solution for MySQL like Vitess (vitess.io) does allow you
to do both cross-shard joins and transactions, but just because you
can doesn’t mean you should. A sharded architecture will perform
best if you design it well and play to its strength, e.g. favoring
single-shard targeted writes within any individual transaction.
Enabling two-phase commit in Vitess to support cross-shard writes is
possible, but will come at a significant performance cost. Whether that tradeoff is worth it differs from application to application, and generally speaking, adjusting the schema/workload is considered the better approach.
It’s better to think of microservices as a design principle than as
a scaling trick. This architecture is more tailored to improving
resilience and flexibility of deployment, by breaking up monolithic
deployments into more loosely coupled, isolated elements. Because
the complexity of managing resources for horizontal sharding align
closely with the challenges of managing resources in a microservices
architecture, Vitess is actually a great fit for a container
orchestration environment and is commonly deployed/managed in
containers using the Vitess Operator for Kubernetes.
In short, horizontally scaling MySQL is both feasible and commonly done, both in microservices architectures, as well as more traditional environments.
As an aside, 10K concurrent requests is relatively vague measure. It’s not abnormal for a well-configured single-server MySQL installation to serve hundreds of thousands of queries per second, so keep in mind that any scaling challenges you might face could also be resolved by optimizing your code, queries, schema and/or MySQL configuration.
I am trying to understand what is the use case of a queue in distributed system.
And also how it scales and how it makes sure it's not a single point of failure in the system?
Any direct answer or a reference to a document is appreciated.
Use case:
I understand that queue is a messaging system. And it decouples the systems that communicate between each other. But, is that the only point of using a queue?
Scalability:
How does the queue scale for high volumes of data? Both read and write.
Reliability:
How does the queue not becoming a single point of failure in the system? Does the queue do a replication, similar to data-storage?
My question is not specified to any particular queue server like Kafka or JMS. Just in general.
Queue is a mental concept, the implementation decides about 1 + 2 + 3
A1: No, it is not the only role -- a messaging seems to be main one, but a distributed-system signalling is another one, by no means any less important. Hoare's seminal CSP-paper is a flagship in this field. Recent decades gave many more options and "smart-behaviours" to work with in designing a distributed-system signalling / messaging services' infrastructures.
A2: Scaling envelopes depend a lot on implementation. It seems obvious that a broker-less queues can work way faster, that a centralised, broker-based, infrastructure. Transport-classes and transport-links account for additional latency + performance degradation as the data-flow volumes grow. BLOB-handling is another level of a performance cliff, as the inefficiencies are accumulating down the distributed processing chain. Zero-copy ( almost ) zero-latency smart-Queue implementations are still victims of the operating systems and similar resources limitations.
A3: Oh sure it is, if left on its own, the SPOF. However, Theoretical Cybernetics makes us safe, as we can create reliable systems, while still using error-prone components. ( M + N )-failure-resilient schemes are thus achievable, however the budget + creativity + design-discipline is the ceiling any such Project has to survive with.
my take:
I would be careful with "decouple" term - if service A calls api on service B, there is coupling since there is a contract between services; this is true even if the communication is happening over a queue, file or fax. The key with queues is that the communication between services is asynchronous. Which means their runtimes are decoupled - from practical point of view, either of systems may go down without affecting the other.
Queues can scale for large volumes of data by partitioning. From clients point of view, there is one queue, but in reality there are many queues/shards and number of shards helps to support more data. Of course sharding a queue is not "free" - you will lose global ordering of events, which may need to be addressed in you application.
A good queue based solution is reliable based on replication/consensus/etc - depends on set of desired properties. Queues are not very different from databases in this regard.
To give you more direction to dig into:
there an interesting feature of queues: deliver-exactly-once, deliver-at-most-once, etc
may I recommend Enterprise Architecture Patterns - https://www.enterpriseintegrationpatterns.com/patterns/messaging/Messaging.html this is a good "system design" level of information
queues may participate in distributed transactions, e.g. you could build something like delete a record from database and write it into queue, and that will be either done/committed or rolledback - another interesting topic to explore
What other stress test cases are there other than finding out the maximum number of users allowed to login into the web application before it slows down the performance and eventually crashing it?
This question is hard to answer thoroughly since it's too broad.
Anyway many stress tests depend on the type and execution flow of your workload. There's an entire subject dedicated (as a graduate course) to queue theory and resources optimization. Most of the things can be summarized as follows:
if you have a resource (be it a gpu, cpu, memory bank, mechanical or
solid state disk, etc..), it can serve a number of users/requests per
second and takes an X amount of time to complete one unit of work.
Make sure you don't exceed its limits.
Some systems can also be studied with a probabilistic approach (Little's Law is one of the most fundamental rules in these cases)
There are a lot of reasons for load/performance testing, many of which may not be important to your project goals. For example:
- What is the performance of a system at a given load? (load test)
- How many users the system can handle and still meet a specific set of performance goals? (load test)
- How does the performance of a system changes over time under a certain load? (soak test)
- When will the system will crash under increasing load? (stress test)
- How does the system respond to hardware or environment failures? (stress test)
I've got a post on some common motivations for performance testing that may be helpful.
You should also check out your web analytics data and see what people are actually doing.
It's not enough to simply simulate X number of users logging in. Find the scenarios that represent the most common user activities (anywhere between 2 to 20 scenarios).
Also, make sure you're not just hitting your cache on reads. Add some randomness / diversity in the requests.
I've seen stress tests where all the users were requesting the same data which won't give you real world results.
HP NonStop systems (previously known as "Tandem") are known for their high availability and reliability, and higher price.
How do Linux or Unix based clusters compare with them, in these respects and others?
On a fault-tolerant machine the fault tolerance is handled directly in hardware and transparent to the application. Programming a cluster requires you to explicitly handle the fault tolerance in the application.
In practice, a clustered application architecture is much more complex to build and error prone than an application built for a fault-tolerant platform such as NonStop. This means that there is a far greater scope for unreliability driven by application bugs, as the London Stock Exchange found out the hard way. They had an incumbent Tandem-based system, which was quite a common architecture for stock exchange trading applications. Their new CEO had the bright idea that Microsoft was the way forward and had a big-5 consultancy build a .Net system based on a cluster of 120 servers.
The problem with clustered applications is that the failures can be correlated. If an application or configuration bug exists in the system it will typically be replicated on all of the nodes. This means that you can get a single situation or event that can take out the whole cluster. The additional complexity of clustered applications makes them more error-prone to develop and deploy, which raises the odds of this happening. A clustered system built on (for example) Linux and J2EE is vulnerable to the same types of failure modes.
IMHO this is a major advantage of older-style mainframe architectures. Several vendors (IBM, HP, DEC and probably several others I can't think of) made fault tolerant systems. The underlying programming model for this type of system is somewhat simpler than a clustered n-tier application server. This means that there is comparatively little to go wrong and for a given amount of effort you can achieve a more reliable system. A surprising number of older architectures are still alive and well and living quite comfortably in their market niches. IBM still sell plenty of Z and I series machines; Unisys still makes the A Series and 2200 series; VMS and NonStop are still alive within HP. The sales of these systems are not all to existing clients - for example a Commercial Underwriting system (GENIUS) runs on the ISeries and is still a market leader in this niche with new rollouts going on as I write this. The application has survived two attempts to rewrite it (1 in in Java and 1 in .Net) that I am aware of and the 'Old School' platform doesn't really seem to be cramping its style.
I wouldn't go shorting any screen-scraper vendors just yet ...
Gray & Reuter's Transaction Processing: Concepts and Techniques is somewhat dry and academic, but has a good treatment of fault-tolerant systems architecture. One of the authors was a key player in the design of Tandem's systems.