(This post is co-authored by Daniel Abadi and Jose Faleiro and cross-posted on Jose's blog)
NoSQL
systems such as MongoDB, Cassandra, HBase, DynamoDB, and Riak have made many
things easier for application developers. They generally have extremely
flexible data models, that reduce the burden of advance prediction of how an
application will change over time. They support a wide variety of data types,
allow nesting of data, and dynamic addition of new attributes. Furthermore, on
the whole, they are relatively easy to install, with far fewer configuration
parameters and dependencies than many traditional database systems.
On the
other hand, their lack of support for traditional atomic transactions is a
major step backwards in terms of ease-of-use for application developers. An
atomic transaction enables a group of writes (to different items in the database)
to occur in an all-or-nothing fashion --- either they will all succeed and be
reflected in the database state, or none of them will. Moreover, in combination
with appropriate concurrency control mechanisms, atomicity guarantees that
concurrent and subsequent transactions either observe all of the completed
writes of an atomic transaction or none of them. Without atomic transactions,
application developers have to write corner-case code to account for cases in
which a group of writes (that are supposed to occur together) have only
partially succeeded or only partially observed by concurrent processes. This
code is error-prone, and requires complex understanding of the semantics of an
application.
At first it
may seem odd that these NoSQL systems, that are so well-known for their
developer-friendly features, should lack such a basic ease-of-use tool as an
atomic transaction. One might have thought that this missing feature is a
simple matter of maturity --- these systems are relatively new and perhaps they
simply haven't yet gotten around to implementing support for atomic
transactions. Indeed, Cassandra's "batch update" feature could be viewed as a
mini-step in this direction (despite the severe constraints on what types of
updates can be placed in a "batch update"). However, as we start to
approach a decade since these systems were introduced, it is clear that there
is a more fundamental reason for the lack of transactional support in these
systems.
Indeed,
there is a deeper reason for their lack of transactional support, and it stems
from their focus on scalability. Most NoSQL systems are designed to scale
horizontally across many different machines, where the data in a database is
partitioned across these machines. The writes in a (general) transaction may
access data in several different partitions (on several different machines).
Such transactions are called "distributed transactions". Guaranteeing
atomicity in distributed transactions requires that the machines that
participate in the transaction coordinate with each other. Each machine must
establish that the transaction can successfully commit on every other machine
involved in the transaction. Furthermore, a protocol is used to ensure that no
machine involved in the transaction will fail before the writes that it was
involved in for that transactions are present in stable storage. This avoids
scenarios where one set of nodes commit a transaction's writes, while another
set of nodes abort or fail before the transaction is complete (which violates
the all-or-nothing guarantee of atomicity).
This
coordination process is expensive, both, in terms of resources, and in terms of
adding latency to database requests. However, the bigger issue is that other
operations are not allowed to read the writes of a transaction until this
coordination is complete, since the all-or-nothing nature of transaction
execution implies that these writes may need to be rolled-back if the
coordination process determines that some of the writes cannot complete and the
transaction must be aborted. The delay of concurrent transactions can cause
further delay of other transactions that have overlapping read- and write-sets
with the delayed transactions, resulting in overall "cloggage" of the
system. The distributed coordination that is required for distributed
transactions thus has significant drawbacks for overall database system
performance --- both in terms of the throughput of transactions per unit time that the system can process, and in terms of the latency of transactions as they get caught up in the cloggage (this cloggage latency often dominates the latency of the transaction coordination protocol itself). Therefore, most NoSQL systems have chosen to disallow general
transactions altogether rather than become susceptible to the performance
pitfalls that distributed transactions can entail.
MongoDB,
Riak, HBase, and Cassandra all provide support for transactions on a single
key. This is because all information associated with a single key is stored on
a single machine (aside from replicas stored elsewhere). Therefore,
transactions on a single key are guaranteed not to involve the types of
complicated distributed coordination described above.
Given that
distributed transactions necessitate distributed coordination, it would seem
that there is a fundamental tradeoff between scalable performance and support
for distributed transactions. Indeed, many practitioners assume that this is
the case. When they set out to build a scalable system, they immediately assume
that they will not be able to support distributed atomic transactions without
severe performance degradation.
This is in
fact completely false. It is very much possible for a scalable system to
support performant distributed atomic transactions.
In a recent paper, we published a new representation of the tradeoffs involved in
supporting atomic transactions in scalable systems. In particular, there exists a three-way
tradeoff between fairness, isolation, and throughput (FIT). A scalable database
system which supports atomic distributed transactions can achieve at most two
out of these three properties. Fairness corresponds to the intuitive notion
that the execution of any given transaction is not deliberately delayed in
order to benefit other transactions.
Isolation provides each transaction with the illusion that it has the
entire database system to itself. In doing so, isolation guarantees that if any
pair of transactions conflict, then one transaction in the pair will always
observe the writes of the other. As a consequence, it alleviates application
developers from the burden of reasoning about complex interleavings of
conflicting transactions' reads and writes. Throughput refers to the ability of
the database to process many concurrent transactions per unit time (without
hiccups in performance due to clogging).
The FIT
tradeoff dictates that there exist three classes of systems that support atomic
distributed transactions:
- Those that guarantee fairness and isolation, but sacrifice throughput,
- Those that guarantee fairness and throughput, but sacrifice isolation, and
- Those that guarantee isolation and throughput, but sacrifice fairness.
In other
words, not only is it possible to build scalable systems with high throughput
distributed transactions, but there actually exist two classes of systems that
can do so: those that sacrifice isolation, and those that sacrifice fairness.
We discuss each of these two alternatives below.
(Latency is not explicitly mentioned in the tradeoff, but systems that give up throughput also give up latency due to cloggage, and systems that give up fairness yield increased latency for those transactions treated unfairly.)
Give up on
isolation
As
described above, the root source of the database system cloggage isn't the
distributed coordination itself. Rather, it is the fact that other transactions
that want to access the data that a particular transaction wrote have to wait
until after the distributed coordination is complete before reading or writing the
shared data. This waiting occurs due to strong isolation, which guarantees that
one transaction in a pair of conflicting must observe the writes of the other.
Since a transaction's writes are not guaranteed to commit until after the
distributed coordination process is complete, concurrent conflicting
transactions cannot make progress for the duration of this coordination.
However,
all of this assumes that it is unacceptable for transactions writes to not be
immediately observable by concurrent conflicting transactions If this
"isolation" requirement is dropped, there is no need for other
transactions to wait until the distributed coordination is complete before
executing and committing.
While
giving up on strong isolation seemingly implies that distributed databases
cannot guarantee correctness (because transactions execute against potentially
stale database state), it turns out that there exists a class of database
constraints that can be guaranteed to hold despite the use of weak isolation
among transactions. For more details on the kinds of guarantees that can hold
on constraints despite weak isolation, Peter Bailis's work on Read Atomic Multi-Partition
(RAMP) transactions provides some great intuition.
Give up on
fairness
The
underlying motivation for giving up isolation in systems is that distributed
coordination extends the duration for which transactions with overlapping data
accesses are unable to make progress. Intuitively, distributed coordination and
isolation mechanisms overlap in time.
This suggests that another way to circumvent the interaction between
isolation techniques and distributed coordination is to re-order distributed
coordination such that its overlap with any isolation mechanism is minimized.
This intuition forms the basis of Isolation-Throughput systems (which give up
fairness). In giving up fairness,
database systems gain the flexibility to pick the most opportune time to pay
the cost of distributed coordination.
For instance, it is possible to perform coordination outside of
transaction boundaries so that the additional time required to do the
coordination does not increase the time that conflicting transactions cannot
run. In general, when the system does not need to guarantee fairness, it can deliberately
prioritize or delay specific transactions in order to benefit overall
throughput.
G-Store is
a good example of an Isolation-Throughput system (which gives up
fairness). G-Store extends a
(non-transactional) distributed key-value store with support for multi-key
transactions. G-Store restricts the
scope of transactions to an application defined set of keys called a KeyGroup.
An application defines KeyGroups dynamically based on the set of keys it
anticipates will be accessed together over the course of some period of time.
Note that the only restriction on transactions is that the keys involved in the
transaction be part of a single KeyGroup. G-Store allows KeyGroups to be
created and disbanded when needed, and therefore effectively provides arbitrary
transactions over any set of keys.
When an
application defines a KeyGroup, G-Store moves the constituent keys from their
nodes to a single leader node. The leader node copies the corresponding
key-value pairs, and all transactions on the KeyGroup are executed on the
leader. Since all the key-value pairs involved in a transaction are stored on a
single node (the leader node), G-Store transactions do not need to execute a
distributed commit protocol during transaction execution.
G-Store
pays the cost of distributed coordination prior to executing transactions. In
order to create a KeyGroup, G-Store executes an expensive distributed protocol
to allow a leader node to take ownership of a KeyGroup, and then move the
KeyGroup's constituent keys to the leader node. The KeyGroup creation protocol
involves expensive distributed coordination, the cost of which is amortized
across the transactions which execute on the KeyGroup.
The key
point is that while G-Store still must perform distributed coordination, this
coordination is done prior to transaction execution --- before the need to be
concerned with isolation from other transactions. Once the distributed
coordination is complete (all the relevant data has been moved to a single
master node), the transaction completes quickly on a single node without
forcing concurrent transactions with overlapping data accesses to wait for
distributed coordination. Hence, G-Store achieves both high throughput and
strong isolation.
However,
the requirement that transactions restrict their scope to a single KeyGroup
favors transactions that execute on keys which have already been grouped. This
is "unfair" to transactions that need to execute on a set of as yet
ungrouped keys. Before such transactions can begin executing, G-Store must
first disband existing KeyGroups to which some keys may belong, and then create
the appropriate KeyGroup --- a process with much higher latency than if the
desired KeyGroup already existed.
Conclusions
The
fundamental reason for the poor performance of conventional distributed
transactions is the fact that the mechanisms for guaranteeing atomicity
(distributed coordination), and isolation overlap in time. The key to enabling
high throughput distributed transactions is to separate these two concerns. This
insight leads to two ways of separating atomicity and isolation mechanisms. The
first option is to weaken isolation such that conflicting transactions can
execute and commit in parallel. The second option is to re-order atomicity and
isolation mechanisms so that they do not overlap in time, and in doing so, give
up fairness during transaction execution.
(Edit: MongoDB and HBase both have (or will soon have) limited support for multi-key transactions as long as those keys are within the same partition. However, hopefully it is clear to the reader that this post is discussing the difficulties of implementing distributed --- cross-partition --- transactions).
Great article Daniel.
ReplyDeleteInteresting you mentioned the idea of weakening isolation.
Relational databases have a way to do exactly that. In MySQL and Oracle for example you can set your isolation level to:
REPEATABLE READ
READ COMMITTED
READ UNCOMMITTED
SERIALIZABLE
And its available for *exactly* the reasons you outline. What's more it has concomitant affects on performance.
Thx for the post!
I half agree. Indeed many traditional database systems have the reduced isolation options you listed. And the main reason for them is to improve throughout at the cost of isolation. However these options existed before distributed database systems became a reality. They are focused on the single-node isolation-throughput tradeoff. In contast, this post talks about the three-way fairness, isolation, throughput tradeoff for distributed transactions.
Delete"Distributed Transactions for Google App Engine" also addresses this problem: http://arxiv.org/html/1106.3325v1
ReplyDeletePer the recommendation of one of our co-authors, Ryan Barrett, the lead on the GAE datastore, we used optimistic concurrency which helps the throughput of concurrent transactions accessing the same data; our algorithm is also tolerant of timeouts. -- Daniel S. Wilkerson
Unfortunately, OCC does not get around the FIT tradeoff. While locking protocols cause cloggage due to transactions waiting for locks, OCC causes cloggage due to transaction aborts. The key thing is that when you extend transactions through coordination across the different partitions, then isolation protocols such as locking and OCC will slow down throughput.
DeleteStill, with proper schema/denormalization OCC is perfect for many use-cases. No transactions at all is the worst you can get (in terms of transactions)...
DeleteI am using a hosted NoSQL database over the Internet. The only reason I am using it is because it is a JSON document store, and I really dont care about transaction isolation levels or distributed transactions. I care that I dont have to pay for a database server license fee and dont have to deal with installing clients because it is all online accessible through the browser. - L Mohan Arun
ReplyDeletewww.mohanarun.com
RavenDB seem to support transactions, not sure what the tradeoffs are, just heard about it recently (probably because of the price ;)
ReplyDeleteZODB does transactions across multiple objects correctly.
ReplyDeletehttp://blogory.org/group/zodb If they are in the same file storage no problem. If they are in different shards, then it uses two-phase commits.
You have spent thousand words just to figure out issue of CAP theorem and have no even reference to it.
ReplyDeleteFIT and CAP are totally different sets of tradeoffs. I have talked about CAP extensively on this blog. Most notably, see my post from a few years ago: http://dbmsmusings.blogspot.com/2010/04/problems-with-cap-and-yahoos-little.html
DeleteI have been following this line of work since reading the "The problems with ACID, and how to fix them without going NoSQL" post in 2010. It has been good to see the published progress on Calvin, and I hope to see the technology gain adoption (whether through Calvin when it is ready or integration into existing solutions) in the coming years.
ReplyDeleteFirst PACELC and now this. Thank you for making it easier to understand and discuss distributed systems.
ReplyDeleteThanks Mark. That means a lot coming from you.
DeleteNot sure how can we sacrifice isolation. If one transaction updates (increments/decrements) two counters, other transaction can come in and set its own values to counters breaking the logic in the first transaction?
ReplyDeleteIt depends on the application. Some applications require stronger isolation guarantees than others. However, please note that when we talk about "sacrificing" isolation, we don't necessarily mean running at zero isolation. We're talking about running at weaker isolation guarantees than full serializability.
DeleteGreat work to help understand distributed transactions! Thank you.
ReplyDeleteI have one question about the definition of "fairness". I am confused by "the execution of any given transaction is not deliberately delayed in order to benefit other transactions". I am not sure whether your definition is really about the fairness. I think there are two concepts. (1) Harm: A is harmed by B, so we say it is unfair for A. (2)Fairness of Harm: A is harmed by B, and B is equally harmed by A. We can say it is fair for both A and B. For example, time sharing in OS scheduling can give fairness, although "the execution of a process is deliberately delayed to benefit other processes".
Thanks for the blog post, very interesting. I'd be keen to hear what you think of other NoSQL solutions such as Datomic.
ReplyDelete