- [Docs] Minor edits

README.md

@@ -25,11 +25,12 @@ for AI workloads both in the cloud and on-prem, speeding up the development, tra
 
 ## Major news ✨
 
-- [2024/08] [dstack 0.18.11: AMD, encryption, and more](https://github.com/dstackai/dstack/releases/tag/0.18.11) (Release)
-- [2024/08] [dstack 0.18.10: Control plane UI](https://github.com/dstackai/dstack/releases/tag/0.18.10) (Release)
-- [2024/07] [dstack 0.18.7: Fleets, RunPod volumes, dstack apply, and more](https://github.com/dstackai/dstack/releases/tag/0.18.7) (Release)
-- [2024/05] [dstack 0.18.4: Google Cloud TPU, and more](https://github.com/dstackai/dstack/releases/tag/0.18.4) (Release)
-- [2024/05] [dstack 0.18.2: On-prem clusters, private subnets, and more](https://github.com/dstackai/dstack/releases/tag/0.18.2) (Release)
+- [2024/10] [dstack 0.18.17: on-prem AMD GPUs, AWS EFA, and more](https://github.com/dstackai/dstack/releases/tag/0.18.17)
+- [2024/08] [dstack 0.18.11: AMD, encryption, and more](https://github.com/dstackai/dstack/releases/tag/0.18.11)
+- [2024/08] [dstack 0.18.10: Control plane UI](https://github.com/dstackai/dstack/releases/tag/0.18.10)
+- [2024/07] [dstack 0.18.7: Fleets, RunPod volumes, dstack apply, and more](https://github.com/dstackai/dstack/releases/tag/0.18.7)
+- [2024/05] [dstack 0.18.4: Google Cloud TPU, and more](https://github.com/dstackai/dstack/releases/tag/0.18.4)
+- [2024/05] [dstack 0.18.2: On-prem clusters, private subnets, and more](https://github.com/dstackai/dstack/releases/tag/0.18.2)
 
 ## Installation
 

docs/docs/concepts/fleets.md

@@ -49,10 +49,11 @@ are both acceptable).
     to the specified parameters.
 
     !!! info "Network"
-        Set `placement` to `cluster` if the nodes should be interconnected
-        (e.g. if you'd like to use them for [multi-node tasks](../reference/dstack.yml/task.md#distributed-tasks)).
-        In that case, `dstack` will provision all nodes in the same backend and region and configure the optimal
-        connectivity via availability zones, placement groups, etc.
+        To ensure the nodes of the fleet are interconnected (e.g., if you'd like to use them for
+        [multi-node tasks](../reference/dstack.yml/task.md#distributed-tasks)), 
+        set `placement` to `cluster`. 
+        In this case, `dstack` will provision all nodes in the same backend and region and configure optimal 
+        inter-node connectivity.
 
         ??? info "AWS"
             `dstack` automatically enables [Elastic Fabric Adapter :material-arrow-top-right-thin:{ .external }](https://aws.amazon.com/hpc/efa/){:target="_blank"}
@@ -62,7 +63,7 @@ are both acceptable).
             `g6.16xlarge`, `g6.24xlarge`, `g6.48xlarge`, `g6.8xlarge`, `gr6.8xlarge`
 
             Currently, only one EFA interface is enabled regardless of the maximum number of interfaces supported by the instance type.
-            This limitation will be resolved in the future.
+            This limitation will be lifted once [this issue :material-arrow-top-right-thin:{ .external }](https://github.com/dstackai/dstack/issues/1804){:target="_blank"} is fixed.
 
     Note that cloud fleets aren't supported for the `kubernetes`, `vastai`, and `runpod` backends.
 

docs/docs/reference/dstack.yml/task.md

@@ -258,9 +258,12 @@ resources:
 </div>
 
 If you run the task, `dstack` first provisions the master node and then runs the other nodes of the cluster.
-All nodes are provisioned in the same region. It is recommended to use a [fleet](../../concepts/fleets.md),
-as in this case all nodes are provisioned into a cluster placement group for better connectivity if the backend
-supports this feature.
+
+??? info "Network"
+    To ensure all nodes are provisioned into a cluster placement group and to enable the highest level of inter-node 
+    connectivity, it is recommended to manually create a  [fleet](../../concepts/fleets.md) before running a task.
+    This won’t be needed once [this issue :material-arrow-top-right-thin:{ .external }](https://github.com/dstackai/dstack/issues/1805){:target="_blank"} 
+    is fixed.
 
 > `dstack` is easy to use with `accelerate`, `torchrun`, and other distributed frameworks. All you need to do
 is pass the corresponding environment variables such as `DSTACK_GPUS_PER_NODE`, `DSTACK_NODE_RANK`, `DSTACK_NODES_NUM`,