Mention monitoring and updating for exchanges (#12953) (#12960)

* Mention monitoring and updating for exchanges

* Fix link syntax...

* Apply suggestions from code review

Co-authored-by: Tyera Eulberg <teulberg@gmail.com>
Co-authored-by: Trent Nelson <trent.a.b.nelson@gmail.com>

* Apply suggestions from code review

Co-authored-by: Tyera Eulberg <teulberg@gmail.com>

* More review comments and word-wrapping

Co-authored-by: Tyera Eulberg <teulberg@gmail.com>
Co-authored-by: Trent Nelson <trent.a.b.nelson@gmail.com>
(cherry picked from commit 87311cce7f)

Co-authored-by: Ryo Onodera <ryoqun@gmail.com>

docs/src/integrations/exchange.md

@@ -7,9 +7,19 @@ exchange.
 
 ## Node Setup
 
-We highly recommend setting up at least two of your own Solana api nodes to
+We highly recommend setting up at least two nodes on high-grade computers/cloud
-give you a trusted entrypoint to the network, allow you full control over how
+instances, upgrading to newer versions promptly, and keeping an eye on service
-much data is retained, and ensure you do not miss any data if one node fails.
+operations with a bundled monitoring tool.
+
+This setup enables you:
+- to have a trusted gateway to the Solana mainnet-beta cluster to get data and
+  submit withdrawal transactions
+- to have full control over how much historical block data is retained
+- to maintain your service availability even if one node fails
+
+Solana nodes demand relatively high computing power to handle our fast blocks
+and high TPS.  For specific requirements, please see
+[hardware recommendations](../running-validator/validator-reqs.md).
 
 To run an api node:
 
@@ -43,12 +53,37 @@ Optional parameters to consider:
 - `--private-rpc` prevents your RPC port from being published for use by other nodes
 - `--rpc-bind-address` allows you to specify a different IP address to bind the RPC port
 
-### Automatic Restarts
+### Automatic Restarts and Monitoring
 
 We recommend configuring each of your nodes to restart automatically on exit, to
 ensure you miss as little data as possible. Running the solana software as a
 systemd service is one great option.
 
+For monitoring, we provide
+[`solana-watchtower`](https://github.com/solana-labs/solana/blob/master/watchtower/README.md),
+which can monitor your validator and detect with the `solana-validator` process
+is unhealthy. It can directly be configured to alert you via Slack, Telegram,
+Discord, or Twillio. For details, run `solana-watchtower --help`.
+
+```bash
+solana-watchtower --validator-identity <YOUR VALIDATOR IDENTITY>
+```
+
+#### New Software Release Announcements
+
+We release new software frequently (around 1 release / week).
+Sometimes newer versions include incompatible protocol changes, which
+necessitate timely software update to avoid errors in processing blocks.
+
+Our official release announcements for all kinds of releases (normal and
+security) are communicated via a discord channel called
+[`#mb-announcement`](https://discord.com/channels/428295358100013066/669406841830244375)
+(`mb` stands for `mainnet-beta`).
+
+Like staked validators, we expect any exchange-operated validators to be updated
+at your earliest convenience within a business day or two after a normal release
+announcement. For security-related releases, more urgent action may be needed.
+
 ### Ledger Continuity
 
 By default, each of your nodes will boot from a snapshot provided by one of your

docs/src/running-validator/validator-reqs.md

@@ -32,6 +32,33 @@ Here are our recommendations for low, medium, and high end machine specification
 | Accounts Drive\(s\) | None                                                  | Samsung 970 Pro 1TB    | 2x Samsung 970 Pro 1TB |                                                                                        |
 | GPU                 | 4x Nvidia 1070 or 2x Nvidia 1080 Ti or 2x Nvidia 2070 | 2x Nvidia 2080 Ti      | 4x Nvidia 2080 Ti      | Any number of cuda-capable GPUs are supported on Linux platforms.                      |
 
+## Virtual machines on Cloud Platforms
+
+While you can run a validator on a cloud computing platform, it may not
+be cost-efficient over the long term.
+
+However, it may be convenient to run non-voting api nodes on VM instances for
+your own internal usage. This use case includes exchanges and services built on
+Solana.
+
+In fact, the offical mainnet-beta API nodes are currently (Oct. 2020) run on GCE
+`n1-standard-32` (32 vCPUs, 120 GB memory) instances with 2048 GB SSD for
+operational convenience.
+
+For other cloud platforms, select instance types with similar specs.
+
+Also note that egress internet traffic usage may turn out to be high,
+especially for the case of running staked validators.
+
+## Docker
+
+Running validator for live clusters (including mainnet-beta) inside Docker is
+not recommended and generally not supported. This is due to concerns of general
+docker's containerzation overhead and resultant performance degradation unless
+specially configured.
+
+We use docker only for development purpose.
+
 ## Software
 
 - We build and run on Ubuntu 18.04. Some users have had trouble when running on Ubuntu 16.04