Set maximum grpc message receive size to 2GiB (#2075)

* Set maximum grpc message receive size to 2GiB

The previous default was 4MiB, which doesn't really work well for
multi-modal models.

* Update to Rust 1.79.0

* Fixup formatting to make PR pass

.github/workflows/tests.yaml

@@ -33,9 +33,9 @@ jobs:
       - name: Install Rust
         uses: actions-rs/toolchain@v1
         with:
-          # Released on: 02 May, 2024
+          # Released on: June 13, 2024
-          # https://releases.rs/docs/1.78.0/
+          # https://releases.rs/docs/1.79.0/
-          toolchain: 1.78.0
+          toolchain: 1.79.0
           override: true
           components: rustfmt, clippy
       - name: Install Protoc

Dockerfile

@@ -1,5 +1,5 @@
 # Rust builder
-FROM lukemathwalker/cargo-chef:latest-rust-1.78 AS chef
+FROM lukemathwalker/cargo-chef:latest-rust-1.79 AS chef
 WORKDIR /usr/src
 
 ARG CARGO_REGISTRIES_CRATES_IO_PROTOCOL=sparse

Dockerfile_amd

@@ -1,5 +1,5 @@
 # Rust builder
-FROM lukemathwalker/cargo-chef:latest-rust-1.78 AS chef
+FROM lukemathwalker/cargo-chef:latest-rust-1.79 AS chef
 WORKDIR /usr/src
 
 ARG CARGO_REGISTRIES_CRATES_IO_PROTOCOL=sparse

Dockerfile_intel

@@ -1,4 +1,4 @@
-FROM lukemathwalker/cargo-chef:latest-rust-1.78 AS chef
+FROM lukemathwalker/cargo-chef:latest-rust-1.79 AS chef
 WORKDIR /usr/src
 
 ARG CARGO_REGISTRIES_CRATES_IO_PROTOCOL=sparse

benchmark/src/app.rs

@@ -497,7 +497,7 @@ fn statis_spans<'a>(data: &[f64], unit: &'static str) -> Vec<Line<'a>> {
                 "Lowest:  {:.2} {unit}",
                 data.iter()
                     .min_by(|a, b| a.total_cmp(b))
-                    .unwrap_or(&std::f64::NAN)
+                    .unwrap_or(&f64::NAN)
             ),
             Style::default().fg(Color::Reset),
         )]),
@@ -506,7 +506,7 @@ fn statis_spans<'a>(data: &[f64], unit: &'static str) -> Vec<Line<'a>> {
                 "Highest: {:.2} {unit}",
                 data.iter()
                     .max_by(|a, b| a.total_cmp(b))
-                    .unwrap_or(&std::f64::NAN)
+                    .unwrap_or(&f64::NAN)
             ),
             Style::default().fg(Color::Reset),
         )]),
@@ -555,17 +555,17 @@ fn latency_throughput_chart<'a>(
     let min_latency: f64 = *latency_iter
         .clone()
         .min_by(|a, b| a.total_cmp(b))
-        .unwrap_or(&std::f64::NAN);
+        .unwrap_or(&f64::NAN);
     let max_latency: f64 = *latency_iter
         .max_by(|a, b| a.total_cmp(b))
-        .unwrap_or(&std::f64::NAN);
+        .unwrap_or(&f64::NAN);
     let min_throughput: f64 = *throughput_iter
         .clone()
         .min_by(|a, b| a.total_cmp(b))
-        .unwrap_or(&std::f64::NAN);
+        .unwrap_or(&f64::NAN);
     let max_throughput: f64 = *throughput_iter
         .max_by(|a, b| a.total_cmp(b))
-        .unwrap_or(&std::f64::NAN);
+        .unwrap_or(&f64::NAN);
 
     // Char min max values
     let min_x = if zoom {

benchmark/src/table.rs

@@ -156,17 +156,17 @@ fn avg_min_max(data: &[f64]) -> (f64, f64, f64) {
     let min = data
         .iter()
         .min_by(|a, b| a.total_cmp(b))
-        .unwrap_or(&std::f64::NAN);
+        .unwrap_or(&f64::NAN);
     let max = data
         .iter()
         .max_by(|a, b| a.total_cmp(b))
-        .unwrap_or(&std::f64::NAN);
+        .unwrap_or(&f64::NAN);
     (average, *min, *max)
 }
 
 fn px(data: &[f64], p: u32) -> f64 {
     let i = (f64::from(p) / 100.0 * data.len() as f64) as usize;
-    *data.get(i).unwrap_or(&std::f64::NAN)
+    *data.get(i).unwrap_or(&f64::NAN)
 }
 
 fn format_value(value: f64, unit: &'static str) -> String {

benchmark/src/utils.rs

@@ -37,7 +37,7 @@ pub(crate) fn percentiles(values: &[f64], pecents: &[i32]) -> BTreeMap<String, f
         .iter()
         .map(|&p| {
             let i = (f64::from(p) / 100.0 * values.len() as f64) as usize;
-            (format!("p{p}"), *values.get(i).unwrap_or(&std::f64::NAN))
+            (format!("p{p}"), *values.get(i).unwrap_or(&f64::NAN))
         })
         .collect()
 }

rust-toolchain.toml

@@ -1,5 +1,5 @@
 [toolchain]
-# Released on: 02 May, 2024
+# Released on: June 13, 2024
-# https://releases.rs/docs/1.78.0/
+# https://releases.rs/docs/1.79.0/
-channel = "1.78.0"
+channel = "1.79.0"
 components = ["rustfmt", "clippy"]

server/text_generation_server/server.py

@@ -240,7 +240,11 @@ def serve(
             interceptors=[
                 ExceptionInterceptor(),
                 UDSOpenTelemetryAioServerInterceptor(),
-            ]
+            ],
+            options=[
+                # Set the maximum possible message length: i32::MAX
+                ("grpc.max_receive_message_length", (1 << 31) - 1)
+            ],
         )
         generate_pb2_grpc.add_TextGenerationServiceServicer_to_server(
             TextGenerationService(model, Cache(), quantize, server_urls), server