![Materialize rows for future processing
let not_west: Vec<String> = not_west_bitset
.iter()
.enumerate()
.filter_map(|(i, &v)| {
if v {
Some(string_vec[i].clone())
} else {
None
}
})
.collect();
let not_west = filter(
&array,
¬_west_bitset
).unwrap();
> Made array of 6666667 Strings not in west
~450 ms
> Made array of 6666667 Strings not in west
~50 ms
+](https://image.slidesharecdn.com/arustyintroductiontoapachearrowandhowitappliestoatimeseriesdatabase-210417112853/85/A-Rusty-introduction-to-Apache-Arrow-and-how-it-applies-to-a-time-series-database-19-320.jpg){kind=icon}
![More efficient encoding (dictionary)
let vb = StringBuilder::new();
let kb = Int8Builder::new();
let mut builder =
StringDictionaryBuilder::new(vb,kb);
(0..NUM_TAGS)
.enumerate()
.for_each(|(i, _)| {
let location = match i % 3 {
0 => "us-east",
1 => "us-midwest",
2 => "us-west",
};
builder.append(location).unwrap();
});
let array = builder.finish();
> total size: 10000688 bytes
10MB
250 ms
+
dictionary
"us-east"
"us-midwest"
"us-west"
Location
0
1
2
0
1
2
0
1
2
[0]
[1]
[2]
[u8]](https://image.slidesharecdn.com/arustyintroductiontoapachearrowandhowitappliestoatimeseriesdatabase-210417112853/85/A-Rusty-introduction-to-Apache-Arrow-and-how-it-applies-to-a-time-series-database-20-320.jpg){kind=icon}
![SIMD Implementation
#[cfg(all(any(target_arch = "x86", target_arch = "x86_64"),
feature = "simd"))]
fn simd_compare_op<T, F>(left: &PrimitiveArray<T>,
right: &PrimitiveArray<T>, op: F) -> Result<BooleanArray>
where
T: ArrowNumericType,
F: Fn(T::Simd, T::Simd) -> T::SimdMask,
{
// use / error checking elided
let null_bit_buffer = combine_option_bitmap(
left.data_ref(), right.data_ref(), len
)?;
let lanes = T::lanes();
let mut result = MutableBuffer::new(
left.len() * mem::size_of::<bool>()
);
let rem = len % lanes;
for i in (0..len - rem).step_by(lanes) {
let simd_left = T::load(left.value_slice(i, lanes));
let simd_right = T::load(right.value_slice(i, lanes));
let simd_result = op(simd_left, simd_right);
T::bitmask(&simd_result, |b| {
result.write(b).unwrap();
});
}
Source: arrow/src/compute/kernels/comparison.rs
if rem > 0 {
let simd_left = T::load(left.value_slice(len - rem, lanes));
let simd_right = T::load(right.value_slice(len - rem, lanes));
let simd_result = op(simd_left, simd_right);
let rem_buffer_size = (rem as f32 / 8f32).ceil() as usize;
T::bitmask(&simd_result, |b| {
result.write(&b[0..rem_buffer_size]).unwrap();
});
}
let data = ArrayData::new(
DataType::Boolean,
left.len(),
None,
null_bit_buffer,
0,
vec![result.freeze()],
vec![],
);
Ok(PrimitiveArray::<BooleanType>::from(Arc::new(data)))
}](https://image.slidesharecdn.com/arustyintroductiontoapachearrowandhowitappliestoatimeseriesdatabase-210417112853/85/A-Rusty-introduction-to-Apache-Arrow-and-how-it-applies-to-a-time-series-database-22-320.jpg){kind=icon}
A Rusty introduction to Apache Arrow and how it applies to a time series database
- 1. A Rusty Introduction to Apache Arrow and how it Applies to a Time Series Database December 9, 2020 Andrew Lamb InfluxData
- 2. IOx Team at InfluxData Query Optimizer / Architect @ Vertica (Columnar Database), Chief Architect @ DataRobot (Machine Learning Platform ) Chief Architect @ Nutonian (Machine Learning Apps XLST JIT Compiler Team at DataPower
- 3. Goals + Outline Goal: ⇒ Arrow is a good basis for a new (time series) Databases ❤ ● Opinions and Perspectives of Databases ● Background on Arrow ● Arrow Examples, in Rust
- 4. Databases -- Trend Towards Specialization Relational Key-Value Timeseries Graph Array / Scientific Document Stream Michael Stonebraker and Ugur Cetintemel. 2005. "One Size Fits All": An Idea Whose Time Has Come and Gone. In Proceedings of the 21st International Conference on Data Engineering (ICDE '05). IEEE Computer Society, USA, 2–11. DOI:https://doi.org/10.1109/ICDE.2005.1 Data Model Deployment Embedded / Edge Cloud Single-Node Hybrid Ecosystem Hadoop Java Json / Javascript AWS GCP Azure Apple Cloud Use Case Transactions Analytics Streaming ...
- 5. … and our new database is … 🎉 InfluxDB IOx - The Future Core of InfluxDB Built with Rust and Arrow
- 6. Analytic Systems (vs Transactional) ● Transactional (OLTP, Key-value stores, etc) ○ Workload is “lookup a record by id”, “update a record”, “keep data durable and consistent” ○ Examples: Oracle, Postgres, Cassandra, DynamoDB, MongoDB, etc etc ● Analytic (OLAP, “Big Data”, etc) ○ Workload: aggregate many rows to get historical view, bulk loads, rarely updated ○ Examples: ClickHouse, MapReduce, Spark, Vertica, Pig, Hive, InfluxDB, etc etc ⇒ Rest of the talk focused on Analytic Databases
- 7. So, you want to build a new database… ? Databases need many features just to look like a database: ● Get Data In and Out ● Store Data and Catalog / Metadata ● Query Store: + Query Language ● Connect: Client API … Before you can invest in what makes your database special
- 8. Implementation timeline for a new Database system Client API In memory storage In-Memory filter + aggregation Durability / persistence Metadata Catalog + Management Query Language Parser Optimized / Compressed storage Execution on Compressed Data Joins! Additional Client Languages Outer Joins Subquery support More advanced analytics Cost based optimizer Out of core algorithms Storage Rearrangement Heuristic Query Planner Arithmetic expressions Date / time Expressions Concurrency Control Data Model / Type System Distributed query execution Resource Management “Lets Build a Database” 🤔 “Ok now this is pretty good” 😐 “Look mom! I have a database!” 😃 Online recovery
- 9. Arrow Project Goals “Build a better open source foundation for data science” 🤔 How is this related to databases? https://arrow.apache.org/
- 10. Arrow == toolkit for a modern analytic databases match tool_needed { File Format (persistence) => Parquet Columnar memory representation => Arrow Arrays Operations (e.g. add, multiply) => Compute Kernels Network transfer => Arrow Flight IPC _ => ... to be continued ... }
- 11. InfluxDB line protocol weather,location=us-east temperature=82,humidity=67 1465839830100400200 weather,location=us-midwest temperature=82,humidity=65 1465839830100400200 weather,location=us-west temperature=70,humidity=54 1465839830100400200 weather,location=us-east temperature=83,humidity=69 1465839830200400200 weather,location=us-midwest temperature=87,humidity=78 1465839830200400200 weather,location=us-west temperature=72,humidity=56 1465839830200400200 weather,location=us-east temperature=84,humidity=67 1465839830300400200 weather,location=us-midwest temperature=90,humidity=82 1465839830400400200 weather,location=us-west temperature=71,humidity=57 1465839830400400200 Line Protocol Tutorial (link) Measurements Tags Fields Timestamp
- 12. IOx Data Model weather,location=us-east temperature=82,humidity=67 1465839830100400200 weather,location=us-midwest temperature=82,humidity=65 1465839830100400200 weather,location=us-west temperature=70,humidity=54 1465839830100400200 weather,location=us-east temperature=83,humidity=69 1465839830200400200 weather,location=us-midwest temperature=87,humidity=78 1465839830200400200 weather,location=us-west temperature=72,humidity=56 1465839830200400200 weather,location=us-east temperature=84,humidity=67 1465839830300400200 weather,location=us-midwest temperature=90,humidity=82 1465839830400400200 weather,location=us-west temperature=71,humidity=57 1465839830400400200 location "us-east" "us-midwest" "us-west" "us-east" "us-midwest" "us-west" "us-east" "us-midwest" "us-west" temperature 82 82 70 83 87 72 84 90 71 humidity 67 65 54 69 78 56 67 82 57 timestamp 2016-06-13T17:43:50.1004002Z 2016-06-13T17:43:50.1004002Z 2016-06-13T17:43:50.1004002Z 2016-06-13T17:43:50.2004002Z 2016-06-13T17:43:50.2004002Z 2016-06-13T17:43:50.2004002Z 2016-06-13T17:43:50.3004002Z 2016-06-13T17:43:50.3004002Z 2016-06-13T17:43:50.3004002Z Table: weather
- 13. Code Examples Thesis: “When writing an analytic database, you will end up implementing the Arrow feature set” (Ecosystem integration is another major benefit of Arrow, subject of a future talk) + * Take performance comparisons with a large grain of salt Compare Plain Rust and Rust using the Arrow library
- 14. Motivating Example “Find the rows that are not in `us-west`”
- 15. Create the Array let string_vec: Vec<String> = (0..NUM_TAGS) .map(|i| { match i % 3 { 0 => "us-east", 1 => "us-midwest", 2 => "us-west", }.into() }) .collect(); let mut builder = StringBuilder::new(NUM_TAGS); (0..NUM_TAGS).enumerate() .for_each(|(i, _)| { let location = match i % 3 { 0 => "us-east", 1 => "us-midwest", 2 => "us-west", }; builder.append_value(location) .unwrap() }); let array = builder.finish(); > created array with 10000000 elements ~600ms > created array with 10000000 elements ~400ms +
- 16. Memory Footprint let size = size_of::<Vec<String>>() + string_vec .iter() .fold(0, |sz, s| { sz + size_of::<String>() + s.len() }); println!("total size: {} bytes", size); println!("total size: {} bytes", array.get_array_memory_size()); > total size: 320000023 bytes ~320 MB * > total size: 149206128 bytes ~150 MB +
- 17. Find Rows != “us-west” let not_west_bitset: Vec<bool> = string_vec .iter() .map(|s| s != "us-west") .collect(); let num_not_west = not_west_bitset .iter() .filter(|&&v| v) .count(); let not_west_bitset = neq_utf8_scalar( &array, "us-west" ).unwrap(); let num_not_west = not_west_bitset .iter() .filter(|v| matches!(v, Some(true))) .count(); > Found 6666667 not in west ~50ms > Found 6666667 not in west ~120ms +
- 18. Find Rows != “us-west” (with null handling) let string_vec: Vec<Option<String>> = ...; let not_west_bitset: Vec<bool> = string_vec .iter() .map(|s| { s.as_ref() .map(|s| s != "us-west") .unwrap_or(false) }) .collect(); let num_not_west = not_west_bitset .iter() .filter(|&&v| v) .count(); + Same as previous > Found 6666667 not in west ~50ms
- 19. Materialize rows for future processing let not_west: Vec<String> = not_west_bitset .iter() .enumerate() .filter_map(|(i, &v)| { if v { Some(string_vec[i].clone()) } else { None } }) .collect(); let not_west = filter( &array, ¬_west_bitset ).unwrap(); > Made array of 6666667 Strings not in west ~450 ms > Made array of 6666667 Strings not in west ~50 ms +
- 20. More efficient encoding (dictionary) let vb = StringBuilder::new(); let kb = Int8Builder::new(); let mut builder = StringDictionaryBuilder::new(vb,kb); (0..NUM_TAGS) .enumerate() .for_each(|(i, _)| { let location = match i % 3 { 0 => "us-east", 1 => "us-midwest", 2 => "us-west", }; builder.append(location).unwrap(); }); let array = builder.finish(); > total size: 10000688 bytes 10MB 250 ms + dictionary "us-east" "us-midwest" "us-west" Location 0 1 2 0 1 2 0 1 2 [0] [1] [2] [u8]
- 21. SIMD Anyone? let output = gt( &left, &right ).unwrap(); + 10 20 17 5 23 5 9 12 4 5 76 2 3 5 2 33 2 1 6 7 8 2 7 2 5 6 7 8 left right output 1 0 1 1 1 0 1 1 0 1 1 0 0 0 > > > >
- 22. SIMD Implementation #[cfg(all(any(target_arch = "x86", target_arch = "x86_64"), feature = "simd"))] fn simd_compare_op<T, F>(left: &PrimitiveArray<T>, right: &PrimitiveArray<T>, op: F) -> Result<BooleanArray> where T: ArrowNumericType, F: Fn(T::Simd, T::Simd) -> T::SimdMask, { // use / error checking elided let null_bit_buffer = combine_option_bitmap( left.data_ref(), right.data_ref(), len )?; let lanes = T::lanes(); let mut result = MutableBuffer::new( left.len() * mem::size_of::<bool>() ); let rem = len % lanes; for i in (0..len - rem).step_by(lanes) { let simd_left = T::load(left.value_slice(i, lanes)); let simd_right = T::load(right.value_slice(i, lanes)); let simd_result = op(simd_left, simd_right); T::bitmask(&simd_result, |b| { result.write(b).unwrap(); }); } Source: arrow/src/compute/kernels/comparison.rs if rem > 0 { let simd_left = T::load(left.value_slice(len - rem, lanes)); let simd_right = T::load(right.value_slice(len - rem, lanes)); let simd_result = op(simd_left, simd_right); let rem_buffer_size = (rem as f32 / 8f32).ceil() as usize; T::bitmask(&simd_result, |b| { result.write(&b[0..rem_buffer_size]).unwrap(); }); } let data = ArrayData::new( DataType::Boolean, left.len(), None, null_bit_buffer, 0, vec![result.freeze()], vec![], ); Ok(PrimitiveArray::<BooleanType>::from(Arc::new(data))) }
- 23. Other things needed in a database Vec<Option<String>> to support nulls Handle other data types with same code Vectorized implementations of filter, aggregate, etc Persist it to storage Send data over the network Ecosystem compatibility ...
- 24. Rust / Arrow Community: Good and Getting better Major Roadmap Items (see also Apache Arrow (Rust) 2.0.0) 1. Support Stable Rust 2. Improved DictionaryArray support and performance 3. Improved compute kernel performance 4. SQL: Joins 5. Parallel CPU-bound operations; Additional platform support (e.g. ARMv8) InfluxData specifically is investing in: 1. Flight IPC 2. Improved Dictionary and Date/Time support 3. Data Fusion (some other tech talk)
- 25. Thank You Find us online Github: https://github.com/influxdata/influxdb_iox Slack: https://influxdata.com/slack It is early days; there are many cool things left to implement And we are hiring (Senior IOx Engineer Job Posting)