fix typo by codespell (#232)

Signed-off-by: Jooho Lee <[email protected]>

Author: Jooho

DEVELOPMENT.md

@@ -58,14 +58,14 @@ $ kubectl --context llm-d-inference-scheduler-dev get service inference-gateway
 # Wait for the LoadBalancer External-IP to become available. The service is accessible over port 80.
 ```
 
-You can now make requests macthing the IP:port of one of the access mode above:
+You can now make requests matching the IP:port of one of the access mode above:
 
 ```bash
 $ curl -s -w '\n' http://<IP:port>/v1/completions -H 'Content-Type: application/json' -d '{"model":"food-review","prompt":"hi","max_tokens":10,"temperature":0}' | jq
 ```
 
 By default the created inference gateway, can be accessed on port 30080. This can
-be overriden to any free port in the range of 30000 to 32767, by running the above
+be overridden to any free port in the range of 30000 to 32767, by running the above
 command as follows:
 
 ```bash
@@ -189,7 +189,7 @@ kubectl config set-context --current --namespace="${NAMESPACE}"
 export HF_TOKEN="<HF_TOKEN>"
 ```
 
-Download the `llm-d-kv-cache-manager` repository (the instllation script and Helm chart to install the vLLM environment):
+Download the `llm-d-kv-cache-manager` repository (the installation script and Helm chart to install the vLLM environment):
 
 ```bash
 cd .. && git clone [email protected]:llm-d/llm-d-kv-cache-manager.git
@@ -206,7 +206,7 @@ make env-dev-kubernetes
 
 This will deploy the entire stack to whatever namespace you chose.
 > [!NOTE]
-> The model and images of each componet can  be replaced. See [Environment Configuration](#environment-configuration) for model settings.
+> The model and images of each component can  be replaced. See [Environment Configuration](#environment-configuration) for model settings.
 
 You can test by exposing the `inference gateway` via port-forward:
 

docs/create_new_filter.md

@@ -17,8 +17,8 @@ Plugins are used to modify llm-d-inference-scheduler's default behavior. Filter
  run in succession to produce the final candidate list which is then evaluated,
  through the process of _scoring_, to select the most appropriate target Pods.
  While llm-d-inference-scheduler comes with several existing filters and
- more are availble in the upstream [Gateway API Inference Extension](https://sigs.k8s.io/gateway-api-inference-extension),
- in some cases it may be desireable to create and deploy custom filtering code to
+ more are available in the upstream [Gateway API Inference Extension](https://sigs.k8s.io/gateway-api-inference-extension),
+ in some cases it may be desirable to create and deploy custom filtering code to
  match your specific requirements.
 
 The filters` main operating function is
@@ -33,9 +33,9 @@ The `Filter` function accepts a `SchedulingContext` (e.g., containing the
  to make scheduling decisions. The function returns a (possibly smaller) array
  of `Pod`s which satisfy the filtering criteria.
 
-## Code walkthough
+## Code walkthrough
 
-The top of the file has the expected Go package and import statments:
+The top of the file has the expected Go package and import statements:
 
 ```go
 package filter
@@ -56,7 +56,7 @@ Specifically, we import the Kubernetes `meta/v1` and `labels` packages to allow
  scheduling related objects) packages.
 
 Next we define the `ByLabels` struct type, along with the relevant fields,
- and a consturctor function.
+ and a constructor function.
 
 ```go
 // ByLabels filters out pods that do not match its label selector criteria

docs/dp.md

@@ -74,7 +74,7 @@ This evolved version removes the requirement for sidecars on the **prefill node*
 ## Worker Selection Logic
 
 - **Decode Worker**:
-  - Prefer longest prefix match / KV cache utilization (depends on avaialble scorers)
+  - Prefer longest prefix match / KV cache utilization (depends on available scorers)
 
 - **Prefill Worker**:
   - High prefix-cache hit rate

pkg/plugins/pre-request/pd_prerequest.go

@@ -74,7 +74,7 @@ func (p *PrefillHeaderHandler) PreRequest(_ context.Context, request *types.LLMR
 		return // prefill profile failed to run or we chose not to run it, no-op in this case
 	}
 
-	// TODO: should the scheme be conifgurable (e.g., https://)?
+	// TODO: should the scheme be configurable (e.g., https://)?
 	prefillURL := "http://" + net.JoinHostPort(prefillProfileRunResult.TargetPod.GetPod().Address, strconv.Itoa(targetPort))
 	request.Headers[prefillPodHeader] = prefillURL
 }

pkg/scheduling/pd/scheduler_test.go

@@ -163,7 +163,7 @@ func TestPDSchedule(t *testing.T) {
 				Prompt:      "12345678901",
 			},
 			input:          []types.Pod{pod1, noRolePod1, noRolePod2},
-			wantRes:        nil, // doesn't mater which pod was selected
+			wantRes:        nil, // doesn't matter which pod was selected
 			unwantedPodIDs: []string{pod1.GetPod().NamespacedName.String()},
 		},
 	}