Move metrics to df-bench

Signed-off-by: Adam Gutglick <[email protected]>

Author: AdamGS

Cargo.lock

@@ -196,7 +196,6 @@ termtree = { version = "0.5" }
 thiserror = "2.0.3"
 tokio = { version = "1.47" }
 tokio-stream = "0.1.17"
-tokio-util = { version = "0.7.16" }
 tpchgen = { version = "2" }
 tpchgen-arrow = { version = "2" }
 tracing = { version = "0.1.41", default-features = false }

Cargo.toml

@@ -24,7 +24,11 @@ datafusion = { workspace = true, features = [
 ] }
 datafusion-common = { workspace = true }
 datafusion-physical-plan = { workspace = true }
+itertools.workspace = true
 object_store = { workspace = true, features = ["aws", "gcp"] }
+opentelemetry.workspace = true
+opentelemetry-otlp.workspace = true
+opentelemetry_sdk.workspace = true
 tokio = { workspace = true, features = ["full"] }
 url = { workspace = true }
 vortex-bench = { workspace = true }

benchmarks/df-bench/Cargo.toml

@@ -1,6 +1,8 @@
 // SPDX-License-Identifier: Apache-2.0
 // SPDX-FileCopyrightText: Copyright the Vortex contributors
 
+pub mod metrics;
+
 use std::sync::Arc;
 
 use datafusion::datasource::file_format::FileFormat;

benchmarks/df-bench/src/lib.rs

@@ -1,6 +1,7 @@
 // SPDX-License-Identifier: Apache-2.0
 // SPDX-FileCopyrightText: Copyright the Vortex contributors
 
+use std::collections::HashMap;
 use std::sync::Arc;
 use std::time::Duration;
 use std::time::SystemTime;
@@ -27,10 +28,8 @@ use opentelemetry_sdk::trace::IdGenerator;
 use opentelemetry_sdk::trace::RandomIdGenerator;
 use opentelemetry_sdk::trace::SpanData;
 use opentelemetry_sdk::trace::SpanExporter;
-use vortex::utils::aliases::hash_map::HashMap;
-
-use crate::Format;
-use crate::utils::GIT_COMMIT_ID;
+use vortex_bench::Format;
+use vortex_bench::utils::GIT_COMMIT_ID;
 
 pub trait MetricsSetExt {
     fn merge_all_with_label(&mut self, other: MetricsSet, labels: &[Label]);

vortex-bench/src/metrics.rs benchmarks/df-bench/src/metrics.rsvortex-bench/src/metrics.rs renamed to benchmarks/df-bench/src/metrics.rs

@@ -43,9 +43,6 @@ itertools = { workspace = true }
 mimalloc = { workspace = true }
 noodles-bgzf = { workspace = true, features = ["async"] }
 noodles-vcf = { workspace = true, features = ["async"] }
-opentelemetry = { workspace = true }
-opentelemetry-otlp = { workspace = true, features = ["trace"] }
-opentelemetry_sdk = { workspace = true }
 parking_lot = { workspace = true }
 parquet = { workspace = true, features = ["async"] }
 rand = { workspace = true }
@@ -58,7 +55,6 @@ tabled = { workspace = true, features = ["std"] }
 target-lexicon = { workspace = true }
 tokio = { workspace = true, features = ["full"] }
 tokio-stream = { workspace = true }
-tokio-util = { workspace = true }
 tpchgen = { workspace = true }
 tpchgen-arrow = { workspace = true }
 tracing = { workspace = true, features = ["max_level_debug"] }

vortex-bench/Cargo.toml

@@ -23,7 +23,6 @@ use vortex_bench::display::DisplayFormat;
 use vortex_bench::display::print_measurements_json;
 use vortex_bench::display::render_table;
 use vortex_bench::measurements::QueryMeasurement;
-use vortex_bench::metrics::MetricsSetExt;
 use vortex_bench::public_bi::FileType;
 use vortex_bench::public_bi::PBI_DATASETS;
 use vortex_bench::public_bi::PBIDataset;
@@ -164,22 +163,23 @@ fn main() -> anyhow::Result<()> {
 
     match args.display_format {
         DisplayFormat::Table => {
-            if args.display_metrics {
-                for (query, format, metric_sets) in metrics {
-                    println!("\nmetrics for query={query}, {format}:");
-                    for (idx, metric_set) in metric_sets.into_iter().enumerate() {
-                        println!("scan[{idx}]:");
-                        for m in metric_set
-                            .timestamps_removed()
-                            .aggregate()
-                            .sorted_for_display()
-                            .iter()
-                        {
-                            println!("{m}");
-                        }
-                    }
-                }
-            }
+            // Temporarily disabled
+            // if args.display_metrics {
+            //     for (query, format, metric_sets) in metrics {
+            //         println!("\nmetrics for query={query}, {format}:");
+            //         for (idx, metric_set) in metric_sets.into_iter().enumerate() {
+            //             println!("scan[{idx}]:");
+            //             for m in metric_set
+            //                 .timestamps_removed()
+            //                 .aggregate()
+            //                 .sorted_for_display()
+            //                 .iter()
+            //             {
+            //                 println!("{m}");
+            //             }
+            //         }
+            //     }
+            // }
             render_table(&mut writer, all_measurements, &args.targets)
         }
         DisplayFormat::GhJson => print_measurements_json(&mut writer, all_measurements),

vortex-bench/src/bin/public_bi.rs

@@ -34,7 +34,6 @@ pub mod engines;
 pub mod fineweb;
 pub mod measurements;
 pub mod memory;
-pub mod metrics;
 pub mod public_bi;
 pub mod random_access;
 pub mod realnest;