Skip to content

Latest commit

 

History

History
751 lines (584 loc) · 37.5 KB

architecture.md

File metadata and controls

751 lines (584 loc) · 37.5 KB

toyDB Architecture

At the highest level, toyDB consists of a cluster of nodes that execute SQL transactions against a replicated state machine. Clients can connect to any node in the cluster and submit SQL statements. It aims to provide linearizability (i.e. strong consistency) and serializability, but falls slightly short as it currently only implements snapshot isolation.

The Raft algorithm is used for cluster consensus, which tolerates the failure of any node as long as a majority of nodes are still available. One node is elected leader, and replicates commands to the others which apply them to local copies of the state machine. If the leader is lost, a new leader is elected and the cluster continues operation. Client commands are automatically forwarded to the leader.

This architecture guide will begin with a high-level overview of node components, before discussing each component from the bottom up. Along the way, we will make note of tradeoffs and design choices.

Node Components

A toyDB node consists of three main components:

  • Storage engine: stores data and manages transactions, on disk and in memory.

  • Raft consensus engine: handles cluster coordination and state machine replication.

  • SQL engine: parses, plans, and executes SQL statements for clients.

These components are integrated in the toyDB server, which handles network communication with clients and other nodes. The following diagram illustrates its internal structure:

toyDB architecture

At the bottom is a simple key/value store, which stores all SQL data. This is wrapped inside an MVCC key/value store that adds ACID transactions. On top of that is a SQL storage engine, providing basic CRUD operations on tables, rows, and indexes. This makes up the node's core storage engine.

The SQL storage engine is wrapped in a Raft state machine interface, allowing it to be managed by the Raft consensus engine. The Raft node receives commands from clients and coordinates with other Raft nodes to reach consensus on an ordered command log. Once commands are committed to the log, they are sent to the state machine driver which applies them to the local state machine.

On top of the Raft engine is a Raft-based SQL storage engine, which implements the SQL storage interface and submits commands to the Raft cluster. This allows the rest of the SQL layer to use the Raft cluster as if it was local storage. The SQL engine manages client SQL sessions, which take SQL queries as text, parse them, generate query plans, and execute them against the SQL storage engine.

Surrounding these components is the toyDB server, which in addition to network communication also handles configuration, logging, and other process-level concerns.

Storage Engine

The storage engine is actually two different storage engines: key/value storage used by the SQL engine, and log-structured storage used by the Raft node. These are both pluggable via the storage_sql and storage_raft configuration options, and have multiple implementations with different characteristics.

The SQL storage engine will be discussed separately in the SQL section.

Key/Value Storage

A key/value storage engine stores arbitrary key/value pairs as binary byte slices, and implements the storage::kv::Store trait:

pub trait Store: Display + Send + Sync {
    /// Deletes a key, or does nothing if it does not exist.
    fn delete(&mut self, key: &[u8]) -> Result<()>;

    /// Flushes any buffered data to the underlying storage medium.
    fn flush(&mut self) -> Result<()>;

    /// Gets a value for a key, if it exists.
    fn get(&self, key: &[u8]) -> Result<Option<Vec<u8>>>;

    /// Iterates over an ordered range of key/value pairs.
    fn scan(&self, range: Range) -> Scan;

    /// Sets a value for a key, replacing the existing value if any.
    fn set(&mut self, key: &[u8], value: Vec<u8>) -> Result<()>;
}

The get, set and delete methods simply read and write key/value pairs, and flush ensures any buffered data is written out to storage (e.g. via the fsync system call). scan iterates over a key/value range in order, a property that is crucial to higher-level functionality (e.g. SQL table scans) and has a couple of important implications:

  • Implementations should store data ordered by key, for performance.

  • Keys should use an order-preserving byte encoding, to allow range scans.

The store itself does not care what keys contain, but the module offers an order-preserving key encoding for use by higher layers. These storage layers often use composite keys made up of several possibly variable-length values (e.g. an index key consists of table, column, and value), and the natural ordering of each segment must be preserved, a property satisfied by this encoding:

  • bool: 0x00 for false, 0x01 for true.
  • Vec<u8>: terminated with 0x0000, 0x00 escaped as 0x00ff.
  • String: like Vec<u8>.
  • u64: Big-endian binary encoding.
  • i64: Big-endian binary encoding, sign bit flipped.
  • f64: Big-endian binary encoding, sign bit flipped if +, all flipped if -.
  • sql::Value: As above, with type prefix 0x00=Null, 0x01=Boolean, 0x02=Float, 0x03=Integer, 0x04=String

The default key/value store is storage::kv::Memory. This is an in-memory B+tree, a search tree variant with multiple keys per node (to make use of cache locality) and values only in leaf nodes. As key/value pairs are added and removed, tree nodes are split, merged, and rotated to keep them balanced and at least half-full.

Although key/value data is stored in memory, toyDB provides durability via the Raft log which is persisted to disk. On startup, the Raft log is replayed to populate the in-memory store.

Key/Value Tradeoffs

In-memory storage: storing key/value data in memory has much better performance and is simpler to implement than on-disk storage, but requires that the data set fits in memory. Replaying the Raft log on startup can also take considerable time for large data sets. However, as toyDB datasets are expected to be small, this is mostly advantageous.

Byte serialization: since the primary storage is in memory, (de)serializing key/value pairs adds significant unnecessary overhead. However, at the outset it was not clear that toyDB would use in-memory storage, and byte slices is a simple interface that can be used regardless of storage medium.

B+tree scans: B+trees often have pointers between neighboring leaf nodes for more efficient range scans, but toyDB's implementation does not. This would complicate the implementation, and the performance benefits are usually not as great in memory where random access latency is low. However, this along with other implementation details cause range scans to be O(log n) rather than O(1) per step.

Key encoding: does not make use of any compression, e.g. variable-length integers, preferring simplicity and correctness.

MVCC Transactions

MVCC (Multi-Version Concurrency Control) is a relatively simple concurrency control mechanism that provides ACID transactions with snapshot isolation without taking out locks or having writes block reads. It also versions all data, allowing querying of historical data.

toyDB implements MVCC at the key/value layer as storage::kv::MVCC, using any storage::kv::Store implementation for underlying storage. begin returns a new transaction, which provides the usual key/value operations such as get, set, and scan. Additionally, it has a commit method which persists the changes and makes them visible to other transactions, and a rollback method which discards them.

When a transaction begins, it fetches the next available transaction ID from Key::TxnNext and increments it, then records itself as an active transaction via Key::TxnActive(id). It also takes a Snapshot, containing the IDs of all other active transactions as of the transaction start, and saves it as Key::TxnSnapshot(id).

Key/value pairs are saved as Key::Record(key, version), where key is the user-provided key and version is the transaction ID which created the record. The visibility of key/value pairs for a transaction is given as follows:

  • For a given user key, do a reverse scan of Key::Record(key, version) starting at the current transaction's ID.

  • Skip any records whose version is in the list of active transaction IDs in the Snapshot.

  • Return the first matching record, if any. This record may be either a Some(value) or a None if the key was deleted.

When writing a key/value pair, the transaction first checks for any conflicts by scanning for a Key::Record(key, version) which is not visible to it. If one is found, a serialization error is returned and the client must retry the transaction. Otherwise, the transaction writes the new record and keeps track of the change as Key::Update(id, key) in case it must roll back later.

When the transaction commits, it simply deletes its Txn::Active(id) record, thus making its changes visible to any subsequent transactions. If the transaction instead rolls back, it iterates over all Key::Update(id, key) entries and removes the written key/value records before removing its Txn::Active(id) entry.

This simple scheme is sufficient to provide ACID transaction guarantees with snapshot isolation: commits are atomic, a transaction sees a consistent snapshot of the key/value store as of the start of the transaction, and any write conflicts result in serialization errors which must be retried.

To satisfy time travel queries, a read-only transaction simply loads the Snapshot entry of a past transaction and applies the same visibility rules as for normal transactions.

MVCC Tradeoffs

Read-only transaction IDs: all transactions, even read-only transactions, are allocated a unique transaction ID. This means that a single standalone SELECT query will result in a write operation to increment the transaction ID counter, which can be expensive.

Serializability: snapshot isolation is not fully serializable, since it exhibits write skew anomalies. This would require serializable snapshot isolation, which was considered unnecessary for a first version - it may be implemented later.

Garbage collection: old MVCC versions are never removed, leading to unbounded disk usage. However, this also allows for complete data history, and simplifies the implementation.

Transaction ID overflow: transaction IDs will overflow after 64 bits, but this is never going to happen with toyDB.

Log-structured Storage

The Raft node needs to keep a log of state machine commands encoded as arbitrary byte slices. This log is mostly append-only, and storing it in a random-access key/value store would be slower and more complex than using a log-structured store purpose-built for this access pattern.

Log stores implement the storage::log::Store trait, a subset of which includes:

pub trait Store: Display + Sync + Send {
    /// Appends a log entry, returning its index.
    fn append(&mut self, entry: Vec<u8>) -> Result<u64>;

    /// Commits log entries up to and including the given index, making them immutable.
    fn commit(&mut self, index: u64) -> Result<()>;

    /// Fetches a log entry, if it exists.
    fn get(&self, index: u64) -> Result<Option<Vec<u8>>>;

    /// Iterates over an ordered range of log entries.
    fn scan(&self, range: Range) -> Scan;

    /// Truncates the log by removing any entries above the given index, and returns the
    /// highest remaining index. Attempting to truncate a committed entry will error.
    fn truncate(&mut self, index: u64) -> Result<u64>;
}

The Raft node appends all received commands to its local log, but only commits entries once they are confirmed by consensus. The local log may need to be truncated, e.g. in the case of a leader change, removing a number of uncommitted entries.

Additionally, the store must be able to store a handful of arbitrary key/value metadata pairs for the Raft node, via set_metadata(key, value) and get_metadata(key) methods.

The default log store in toyDB is storage::log::Hybrid, which stores uncommitted entries in memory and committed entries on disk. This allows the log to be written append-only and in order, giving very good performance both for writes and bulk reads. The number of uncommitted entries is also generally small since consensus is generally fast.

New log entries are kept in a VecDeque (double-ended queue) until they are committed. On commit, entries are appended to the file with a u32 length prefix, and the file is fsynced (if enabled). Entry positions are kept in an in-memory HashMap keyed by entry index, for retrieval, and this map is rebuilt on startup by scanning the log file.

Metadata key/value pairs are kept in an in-memory HashMap and the entire hashmap is written to a separate file on every write.

Log Tradeoffs

Startup log scan: scanning the entire file on startup to build the entry index can be time-consuming, and the index requires a bit of memory. However, this avoids having to maintain separate index storage, which could be expensive to fsync, and data sets are expected to be small.

Metadata storage: metadata key/value pairs should be stored in e.g. an on-disk B-tree key/value store, but toyDB current does not have such a store. However, the number of metadata items is very small - specifically 1: the current Raft term/vote tuple.

Memory buffering: buffering uncommitted entries in memory may require a lot of memory if consensus halts, e.g. due to loss of quorum. However, for toyDB use-cases this is not a major problem, and it avoid having to do additional (possibly random) disk IO, greatly improving performance.

Garbage collection: there is no garbage collection of old log entries, so the log will grow without bound. However, this is a necessity since the the default toyDB configuration uses in-memory key/value storage by default and there is no other durable storage.

Raft Consensus Engine

The Raft consensus protocol is explained well in the original Raft paper, and will not be repeated here - refer to it for details. toyDB's implementation follows the paper fairly closely.

The Raft node raft::Node is the core of the implementation, a finite state machine with enum variants for the node roles: leader, follower, and candidate. This enum wraps the RoleNode struct, which contains common node functionality and is generic over the specific roles Leader, Follower, and Candidate that implement the Raft protocol.

Nodes are initialized with an ID and a list of peer IDs, and communicate by passing raft::Message messages. Inbound messages are received via Node.step() calls, and outbound messages are sent via an mpsc channel. Nodes also use a logical clock to keep track of e.g. election timeouts and heartbeats, and the clock is ticked at regular intervals via Node.tick() calls. These methods are synchronous and may cause state transitions, e.g. changing a candidate into a leader when it receives the winning vote.

Nodes have a command log raft::Log, using a storage::log::Store for storage. Leaders receive client commands via request messages, replicate them to peers, and commit the commands to the log subject to consensus. Once a command is committed, is it applied to the state machine asynchronously.

The Raft-managed state machine (i.e. the SQL storage engine) implements the raft::State trait and is given to the node on initialization. The state machine driver raft::Driver has ownership of the state machine, and runs in a separate thread (or rather, a Tokio task) receiving instructions via an mpsc channel - this avoids long-running commands blocking the main Raft node from responding to messages.

In addition to applying state machine commands, the driver also responds to client requests via an outbound mpsc channel. When the leader receives a state mutation request from a client, it not only appends the command to its log, but it also tells the driver that the client is to be notified with the result once the command is applied. When the leader receives a state query request, the state driver is notified about the query before the leader asks all peers to confirm that it is still the leader (required to satisfy linearizability). The confirmations are passed to the state machine driver, and once a majority vote is received the query is executed against the state machine and the result returned to the client.

The actual network communication is handled by the server process, which will be described in a separate section.

Raft Tradeoffs

Single-threaded state: all state operations run in a single thread on the leader, preventing horizontal scalability. Improvements here would require running multiple sharded Raft clusters, which is out of scope for the project.

Log replication: only the simplest form of Raft log replication is implemented, without state snapshots or rapid log replay. Lagging nodes will be very slow to catch up.

Cluster resizing: the Raft cluster consists of a static set of nodes given at startup, resizing it requires a complete cluster restart.

SQL Engine

The SQL engine builds on Raft and MVCC to provide a SQL interface to clients. Logically, the life of a SQL query is as follows:

Query → Lexer → Parser → Planner → Optimizer → Executor → Storage Engine

We'll begin by looking at the basic SQL type and schema systems, as well as the SQL storage engine and its session interface. Then we'll switch sides and look at how a query is executed, starting at the front with the parser and following it until it's executed against the SQL storage engine, completing the chain.

Types

toyDB has a very simple type system, with the sql::DataType enum specifying the available data types: Boolean, Integer, Float, and String.

The sql::Value enum represents a specific value using Rust's native type system, e.g. an integer value is Value::Integer(i64). This enum also specifies comparison, ordering, and formatting of values. The special value Value::Null represents an unknown value of unknown type, following the rules of three-valued logic.

Values can be grouped into a Row, which is an alias for Vec<Value>. The type Rows is an alias for a fallible row iterator, and Column is a result column containing a name.

Expressions sql::Expression represent operations on values. For example, (1 + 2) * 3 is represented as:

Expression::Multiply(
    Expression::Add(
        Expression::Constant(Value::Integer(1)),
        Expression::Constant(Value::Integer(2)),
    ),
    Expression::Constant(Value::Integer(3)),
)

Calling evaluate() on the expression will recursively evaluate it, returning Value::Integer(9).

Schemas

The schema defines the tables sql::Table and columns sql::Column in a toyDB database. Tables have a name and a list of columns, while a column has several attributes such as name, data type, and various constraints. They also have methods to validate rows and values, e.g. to make sure a value is of the correct type for a column or to enforce referential integrity.

The schema is stored and managed with sql::Catalog, a trait implemented by the SQL storage engine:

pub trait Catalog {
    /// Creates a new table.
    fn create_table(&mut self, table: &Table) -> Result<()>;

    /// Deletes a table, or errors if it does not exist.
    fn delete_table(&mut self, table: &str) -> Result<()>;

    /// Reads a table, if it exists.
    fn read_table(&self, table: &str) -> Result<Option<Table>>;

    /// Iterates over all tables.
    fn scan_tables(&self) -> Result<Tables>;
}

Schema Tradeoffs

Single database: only a single, unnamed database is supported per toyDB cluster. This is sufficient for toyDB's use-cases, and simplifies the implementation.

Schema changes: schema changes other than creating or dropping tables is not supported. This avoids complicated data migration logic, and allows using table/column names as storage identifiers (since they can never change) without any additional indirection.

Storage

The SQL storage engine trait is sql::Engine:

pub trait Engine: Clone {
    type Transaction: Transaction;

    /// Begins a transaction in the given mode.
    fn begin(&self, mode: Mode) -> Result<Self::Transaction>;

    /// Resumes an active transaction with the given ID.
    fn resume(&self, id: u64) -> Result<Self::Transaction>;

    /// Begins a SQL session for executing SQL statements.
    fn session(&self) -> Result<Session<Self>> {
        Ok(Session { engine: self.clone(), txn: None })
    }
}

The main use of the trait is to dispense sql::Session instances, individual client sessions which execute SQL queries submitted as plain text and track transaction state. The actual storage engine functionality is exposed via the sql::Transaction trait, representing an ACID transaction providing basic CRUD (create, read, update, delete) operations for tables, rows, and indexes:

pub trait Transaction: Catalog {
    /// Commits the transaction.
    fn commit(self) -> Result<()>;
    /// Rolls back the transaction.
    fn rollback(self) -> Result<()>;

    /// Creates a new table row.
    fn create(&mut self, table: &str, row: Row) -> Result<()>;
    /// Deletes a table row.
    fn delete(&mut self, table: &str, id: &Value) -> Result<()>;
    /// Reads a table row, if it exists.
    fn read(&self, table: &str, id: &Value) -> Result<Option<Row>>;
    /// Scans a table's rows, optionally filtering by the given predicate expression.
    fn scan(&self, table: &str, filter: Option<Expression>) -> Result<Scan>;
    /// Updates a table row.
    fn update(&mut self, table: &str, id: &Value, row: Row) -> Result<()>;

    /// Reads an index entry, if it exists.
    fn read_index(&self, table: &str, column: &str, value: &Value) -> Result<HashSet<Value>>;
    /// Scans a column's index entries.
    fn scan_index(&self, table: &str, column: &str) -> Result<IndexScan>;
}

The main SQL storage engine implementation is sql::engine::KV, which is built on top of an MVCC key/value store and its transaction functionality.

The Raft SQL storage engine sql::engine::Raft uses a Raft API client raft::Client to submit state machine commands specified by the enums Mutation and Query to the local Raft node. It also provides a Raft state machine sql::engine::raft::State which wraps a regular sql::engine::KV SQL storage engine and applies state machine commands to it. Since the Raft SQL engine implements the sql::Engine trait, it can be used interchangably with the local storage engine.

Storage Tradeoffs

Raft result streaming: result streaming is not implemented for Raft commands, so the Raft SQL engine must buffer the entire result set in memory and serialize it before returning it to the client - particularly expensive for table scans. Implementing streaming in Raft was considered out of scope for the project.

Parsing

The SQL session sql::Session takes plain-text SQL queries via execute() and returns the result. The first step in this process is to parse the query into an abstract syntax tree (AST) which represents the query semantics. This happens as follows:

SQL → Lexer → Tokens → Parser → AST

The lexer sql::Lexer takes a SQL string, splits it into pieces, and classifies them as tokens sql::Token. It does not care about the meaning of the tokens, but removes whitespace and tries to figure out if something is a number, string, keyword, and so on. It also does some basic pre-processing, such as interpreting string quotes, checking number formatting, and rejecting unknown keywords.

For example, the following input string results in the listed tokens, even though the query is invalid:

3.14 +UPDATE 'abc'Token::Number("3.14") Token::Plus Token::Keyword(Keyword::Update) Token::String("abc")

The parser sql::Parser iterates over the tokens generated by the lexer, interprets them, and builds an AST representing the semantic query. For example, SELECT name, 2020 - birthyear AS age FROM people results in the following AST:

ast::Statement::Select{
    select: vec![
        (ast::Expression::Field(None, "name"), None),
        (ast::Expression::Operation(
            ast::Operation::Subtract(
                ast::Expression::Literal(ast::Literal::Integer(2020)),
                ast::Expression::Field(None, "birthyear"),
            )
        ), Some("age")),
    ],
    from: vec![
        ast::FromItem::Table{name: "people", alias: None},
    ],
    where: None,
    group_by: vec![],
    having: None,
    order: vec![],
    offset: None,
    limit: None,
}

The parser will interpret the SQL syntax, determining the type of query and its parameters, returning an error for any invalid syntax. However, it has no idea if the table people actually exists, or if the field birthyear is an integer - that is the job of the planner.

Notably, the parser also parses expressions, such as 1 + 2 * 3. This is non-trivial due to precedence rules, i.e. 2 * 3 should be evaluated first, but not if there are parentheses around (1 + 2). The toyDB parser uses the precedence climbing algorithm for this. Also note that AST expressions are different from SQL engine expressions, and do not map one-to-one. This is clearest in the case of function calls, where the parser does not know (or care) if a given function exists, it just parses a function call as an arbitrary function name and arguments. The planner will translate this into actual expressions that can be evaluated.

Planning

The SQL planner sql::Planner takes the AST generated by the parser and builds a SQL execution plan sql::Plan, which is an abstract representation of the steps necessary to execute the query. For example, the following shows a simple query and corresponding execution plan, formatted as EXPLAIN output:

SELECT id, title, rating * 100 FROM movies WHERE released > 2010 ORDER BY rating DESC;

Order: rating desc
└─ Projection: id, title, rating * 100
   └─ Filter: released > 2010
      └─ Scan: movies

The plan nodes sql::Node in a query execution plan represent a relational algebra operator, where the output from one node flows as input into the next. In the example above, the query first does a full table scan of the movies table, then applies the filter released > 2010 to the rows, before projecting (formatting) the result and sorting it by rating.

Most of the planning is fairly straightforward, translating AST nodes to plan nodes and expressions. The trickiest part is resolving table and column names to result column indexes across multiple layers of aliases, joins, and projections - this is handled with sql::plan::Scope, which keeps track of what names are visible to the node being built and maps them to column indexes. Another challenge is aggregate functions, which are implemented as a pre-projection of function arguments and grouping/hidden columns, then an aggregation node, and finally a post-projection evaluating the final aggregate expressions - like in the following example:

SELECT   g.name AS genre, MAX(rating * 100) - MIN(rating * 100)
FROM     movies m JOIN genres g ON m.genre_id = g.id
WHERE    m.released > 2000
GROUP BY g.id, g.name
HAVING   MIN(rating) > 7 
ORDER BY g.id ASC;

Projection: #0, #1
└─ Order: g.id asc
   └─ Filter: #2 > 7
      └─ Projection: genre, #0 - #1, #2, g.id
         └─ Aggregation: maximum, minimum, minimum
            └─ Projection: rating * 100, rating * 100, rating, g.id, g.name
               └─ Filter: m.released > 2000
                  └─ NestedLoopJoin: inner on m.genre_id = g.id
                     ├─ Scan: movies as m
                     └─ Scan: genres as g

The planner generates a very naïve execution plan, primarily concerned with producing one that is correct but not necessarily fast. This means that it will always do full table scans, always use nested loop joins, and so on. The plan is then optimized by a series of optimizers implementing sql::Optimizer:

  • ConstantFolder: pre-evaluates constant expressions to avoid having to re-evaluate them for each row.

  • FilterPushdown: pushes filters deeper into the query to reduce the number of rows evaluated by each node, e.g. by pushing single-table predicates all the way to the table scan node such that filtered nodes won't have to go across the Raft layer.

  • IndexLookup: transforms table scans into primary key or index lookups where possible.

  • NoopCleaner: attempts to remove noop operations, e.g. filter nodes that evaluate to a constant TRUE value.

  • JoinType: transforms nested loop joins into hash joins for equijoins (equality join predicate).

Optimizers make heavy use of boolean algebra to transform expressions into forms that are more convenient to work with. For example, partial filter pushdown (e.g. across join nodes) can only push down conjunctive clauses (i.e. AND parts), so expressions are converted into conjunctive normal form first such that each part can be considered separately.

Below is an example of a complex optimized plan where table scans have been replaced with key and index lookups, filters have been pushed down into scan nodes, and nested loop joins have been replaced by hash joins. It fetches science fiction movies released since 2000 by studios that have released any movie with a rating of 8 or more:

SELECT   m.id, m.title, g.name AS genre, m.released, s.name AS studio
FROM     movies m JOIN genres g ON m.genre_id = g.id,
         studios s JOIN movies good ON good.studio_id = s.id AND good.rating >= 8
WHERE    m.studio_id = s.id AND m.released >= 2000 AND g.id = 1
ORDER BY m.title ASC;

Order: m.title asc
└─ Projection: m.id, m.title, g.name, m.released, s.name
   └─ HashJoin: inner on m.studio_id = s.id
      ├─ HashJoin: inner on m.genre_id = g.id
      │  ├─ Filter: m.released > 2000 OR m.released = 2000
      │  │  └─ IndexLookup: movies as m column genre_id (1)
      │  └─ KeyLookup: genres as g (1)
      └─ HashJoin: inner on s.id = good.studio_id
         ├─ Scan: studios as s
         └─ Scan: movies as good (good.rating > 8 OR good.rating = 8)

Planning Tradeoffs

Type checking: expression type conflicts are only detected at evaluation time, not during planning.

Execution

Every SQL plan node has a corresponding executor, implementing the sql::Executor trait:

pub trait Executor<T: Transaction> {
    /// Executes the executor, consuming it and returning a result set
    fn execute(self: Box<Self>, txn: &mut T) -> Result<ResultSet>;
}

Executors are given a sql::Transaction to access the SQL storage engine, and return a sql::ResultSet with the query result. Most often, the result is of type sql::ResultSet::Query containing a list of columns and a row iterator. Most executors contain other executors that they use as inputs, for example the Filter executor will often have a Scan executor as a source:

pub struct Filter<T: Transaction> {
    source: Box<dyn Executor<T>>,
    predicate: Expression,
}

Calling execute on a sql::Plan will build and execute the root node's executor, which in turn will recursively call execute on its source executors (if any) and process their results. Executors typically augment the source's returned row iterator using Rust's Iterator functionality, e.g. by calling filter() on it and returning a new iterator. The entire execution engine thus works in a streaming fashion and leverages Rust's zero-cost iterator abstractions.

Finally, the root ResultSet is returned to the client.

Server

The toyDB Server manages network traffic for the Raft and SQL engines, using the Tokio async executor. It opens TCP listeners on port 9605 for SQL clients and 9705 for Raft peers, both using length-prefixed Bincode-encoded message passing via Serde-encoded Tokio streams as a protocol.

The Raft server is split out to raft::Server, which runs a main event loop routing Raft messages between the local Raft node, state machine driver, TCP peers, and local state machine clients (i.e. the Raft SQL engine wrapper), as well as ticking the Raft logical clock at regular intervals. It spawns separate Tokio tasks that maintain outbound TCP connections to all Raft peers, while internal communication happens via mpsc channels.

The SQL server spawns a new Tokio task for each SQL client that connects, running a separate SQL session from the SQL storage engine on top of Raft. It communicates with the client by passing server::Request and server::Response messages that are translated to sql::Session calls.

The main toydb binary simply initializes a toyDB server based on command-line arguments and configuration files, and then runs it via the Tokio runtime.

Server Tradeoffs

Security: all network traffic is unauthenticated an in plaintext, as security was considered out of scope for the project.

Client

The toyDB Client provides a simple API for interacting with a server, mainly by executing SQL statements via execute() returning sql::ResultSet. It also has the convenience method with_txn(), taking a closure that executes a series of SQL statements while automatically catching and retrying serialization errors.

There is also client::Pool, which manages a set of pre-connected clients that can be retrieved for running short-lived queries in a multi-threaded application without incurring connection setup costs.

The toysql command-line client is a simple REPL client that connects to a server using the toyDB Client and continually prompts the user for a SQL query to execute, displaying the returned result.