diff --git a/PLAN.md b/PLAN.md
deleted file mode 100644
index c4291ee41e..0000000000
--- a/PLAN.md
+++ /dev/null
@@ -1,20 +0,0 @@
-# jts-agentic-workflow
-
-## Source
-
-- **Project**: jts-agentic-workflow
-- **Branch**: `docs-v2-jts-agentic-workflow`
-- **Created**: 2026-02-10
-
-## Objective
-
-<!-- What are we trying to accomplish? -->
-
-## Tasks
-
-- [ ] Task 1
-- [ ] Task 2
-
-## Notes
-
-<!-- Research, decisions, blockers -->
diff --git a/content/influxdb3/enterprise/performance-preview/_index.md b/content/influxdb3/enterprise/performance-preview/_index.md
new file mode 100644
index 0000000000..fc788582e4
--- /dev/null
+++ b/content/influxdb3/enterprise/performance-preview/_index.md
@@ -0,0 +1,331 @@
+---
+title: Performance upgrade preview
+seotitle: Performance upgrade preview for InfluxDB 3 Enterprise
+description: >
+  Preview performance upgrades in InfluxDB 3 Enterprise with improved
+  single-series query performance, consistent resource usage, wide-and-sparse
+  table support, column families, and bulk data export.
+menu:
+  influxdb3_enterprise:
+    name: Performance upgrade preview
+weight: 12
+influxdb3/enterprise/tags: [storage, performance, beta, preview]
+related:
+  - /influxdb3/enterprise/get-started/setup/
+  - /influxdb3/enterprise/performance-preview/configure/
+  - /influxdb3/enterprise/performance-preview/monitor/
+  - /influxdb3/enterprise/admin/performance-tuning/
+---
+
+> [!Warning]
+> #### Performance preview beta
+> The performance upgrade preview is available to {{% product-name %}} Trial
+> and Commercial users as a beta. These features are subject to breaking changes
+> and **should not be used for production workloads**.
+>
+> To share feedback on this preview, see [Support and feedback options](#bug-reports-and-feedback).
+> Your feedback on stability
+> and performance at scale helps shape the future of InfluxDB 3.
+
+## What is the performance upgrade preview?
+
+{{% product-name %}} includes a preview of major upgrades to the
+storage layer that improve how data is written, stored, compressed, compacted,
+and queried.
+These upgrades touch every layer of the storage path—from a new on-disk file
+format to how fields are organized into column families and how compaction
+manages resources.
+
+## Why these upgrades
+
+The existing InfluxDB 3 storage layer uses [Apache Parquet](https://parquet.apache.org/)
+and is optimized for analytical workloads.
+Customers running high-cardinality, wide-schema, and query-intensive workloads
+need better single-series query performance, more predictable resource usage,
+and the schema flexibility that made InfluxDB v1 and v2 popular.
+These upgrades extend the storage layer to support those workloads while
+maintaining full compatibility with InfluxDB 3's data model and query languages.
+
+Key improvements include:
+
+- **Faster single-series queries**: Single-digit millisecond response times
+  for highly selective time-series queries.
+- **Consistent resource usage**: Bounded CPU and memory during persistence
+  and compaction, eliminating spikes during heavy ingestion or compaction bursts.
+- **Wide-and-sparse table support**: Schemas with up to millions of columns
+  and dynamic schema evolution without expensive rewrites.
+- **Column families**: Group related fields for efficient compression and I/O,
+  so queries only read the data they need.
+- **Bulk data export**: Export compacted data as Parquet files for use with
+  external tools.
+- **Automatic Parquet upgrade**: Seamlessly migrate existing data with
+  hybrid query mode during the transition.
+
+## Enable the preview
+
+Include the `--use-pacha-tree` flag in your
+[`influxdb3 serve` startup command](/influxdb3/enterprise/get-started/setup/):
+
+{{< code-callout "--use-pacha-tree" >}}
+```bash
+influxdb3 serve \
+  --node-id host01 \
+  --cluster-id cluster01 \
+  --object-store file \
+  --data-dir ~/.influxdb3 \
+  --use-pacha-tree
+```
+{{< /code-callout >}}
+
+You can also enable the preview with an environment variable:
+
+```bash
+export INFLUXDB3_ENTERPRISE_USE_PACHA_TREE=true
+influxdb3 serve ...
+```
+
+The `--use-pacha-tree` flag exposes additional configuration options prefixed
+with `--pt-`.
+See [Configure the preview](/influxdb3/enterprise/performance-preview/configure/)
+for tuning options, or
+[Monitor the preview](/influxdb3/enterprise/performance-preview/monitor/)
+for system tables and telemetry.
+
+> [!Warning]
+> #### Existing clusters with Parquet data
+>
+> On clusters with existing Parquet data, enabling `--use-pacha-tree`
+> **automatically converts Parquet files to `.pt` format** on startup, which
+> consumes additional CPU and memory while the migration runs.
+> Queries continue to work normally during this period.
+> See [Upgrade from Parquet](#upgrade-from-parquet) for details.
+>
+> For the beta, we recommend enabling the preview with a fresh cluster in a
+> staging or test environment first.
+
+## What's changed
+
+These upgrades touch every layer of the storage path—from the on-disk file
+format to how data is compressed, organized, and compacted.
+
+### New file format
+
+Data is stored in a new columnar file format (`.pt` files) optimized for
+time-series workloads.
+All data within a file is sorted by column family key,
+[series key](/influxdb3/enterprise/reference/glossary/#series-key), and
+timestamp, which enables efficient compaction, querying, and filtering.
+
+The format uses type-specific compression algorithms that adapt to data
+characteristics—delta-delta RLE for timestamps, Gorilla encoding for floats,
+dictionary encoding for low-cardinality strings, and more—typically
+achieving 5-20x compression ratios.
+
+### Column families
+
+Column families let you group related fields together so that queries only read
+the data they need.
+Fields in the same family are stored together on disk.
+For wide tables with hundreds of fields, this dramatically reduces I/O.
+
+When writing [line protocol](/influxdb3/enterprise/reference/line-protocol/), use the `::` (double-colon) delimiter in field
+names to assign fields to a family.
+The portion before `::` is the family name; everything after is the field name.
+
+```txt
+metrics,host=sA cpu::usage_user=55.2,cpu::usage_sys=12.1,cpu::usage_idle=32.7 1000000000
+metrics,host=sA mem::free=2048i,mem::used=6144i,mem::cached=1024i 1000000000
+metrics,host=sA disk::read_bytes=50000i,disk::write_bytes=32000i 1000000000
+```
+
+This creates three column families:
+
+| Family | Fields |
+|:-------|:-------|
+| `cpu` | `usage_user`, `usage_sys`, `usage_idle` |
+| `mem` | `free`, `used`, `cached` |
+| `disk` | `read_bytes`, `write_bytes` |
+
+When a query references only `mem::free`, the storage layer reads only the
+`mem` family block and skips `cpu` and `disk` data entirely.
+
+> [!Note]
+> Only the first `::` is significant.
+> A field name like `a::b::c` creates family `a` with field `b::c`.
+
+Fields written without `::` are assigned to auto-generated families (named
+`__0`, `__1`, etc.), each holding up to 100 fields.
+Explicit family names are an excellent way to optimize performance with known
+workloads, but they're not required to achieve good results.
+
+### Bounded compaction
+
+Incoming writes are buffered in the WAL, flushed to snapshots, and then merged
+into [Gen0 files](/influxdb3/enterprise/performance-preview/configure/#gen0).
+The upgraded storage layer organizes compacted data into 24-hour UTC windows
+and progresses Gen0 files through four [compaction levels (L1 through L4)](/influxdb3/enterprise/performance-preview/configure/#l1-l4-level-tuning).
+Compaction runs continuously in the background with a byte-based memory budget
+(default: 50% of system RAM), so it never causes resource spikes.
+
+Old files are cleaned up after a cooldown period, ensuring query replicas have
+time to see new checkpoints before old data is removed.
+Failures are automatically retried, and the system is designed to be
+self-healing for transient issues.
+
+## Upgrade from Parquet
+
+Existing clusters with Parquet data can upgrade with zero manual migration.
+The upgrade is fully automatic and occurs on initial startup.
+
+When you restart a cluster with `--use-pacha-tree`, the system:
+
+1. Detects existing Parquet data and enters hybrid mode.
+2. Clears the legacy WAL on ingest nodes and streams Parquet files through a
+   conversion pipeline.
+3. Integrates converted files into the new storage format through compaction.
+4. Automatically transitions once all data is migrated.
+
+During hybrid mode, queries merge results from both the legacy and upgraded
+storage layers.
+If there is a conflict (same series key and timestamp), the upgraded data takes
+precedence.
+
+### Monitor upgrade progress
+
+Use system tables to track upgrade status:
+
+```sql
+-- Per-node upgrade status
+SELECT * FROM system.upgrade_parquet_node
+
+-- Per-file migration progress
+SELECT * FROM system.upgrade_parquet
+```
+
+### Configure upgrade behavior
+
+| Option | Description | Default |
+|:-------|:------------|:--------|
+| `--pt-disable-hybrid-query` | Disable hybrid query mode. Queries return only data from the upgraded storage layer, even during migration. | `false` |
+| `--pt-upgrade-poll-interval` | Polling interval for upgrade status monitoring. | `5s` |
+
+## Downgrade to Parquet
+
+If you need to revert from the performance preview back to standard Parquet
+storage, use the `influxdb3 downgrade-to-parquet` command.
+This command updates the catalog and deletes all PachaTree-specific files from
+object storage.
+
+> [!Note]
+> #### Downgrade impacts
+>
+> The downgrade deletes all `.pt` files, including data written
+> after the upgrade.
+> **Only data that existed before the upgrade (original Parquet files) is preserved.**
+> You can re-enable the preview later by restarting with `--use-pacha-tree`.
+
+### Before you downgrade
+
+1. **Stop all nodes** in the cluster before running the downgrade command.
+   The command checks for running nodes and refuses to proceed if any are active.
+
+   ```bash
+   influxdb3 stop node --node-id <NODE_ID>
+   ```
+
+2. **Verify table compatibility.**
+   The downgrade validates that all tables can be represented in Parquet format.
+   Tables that exceed the Parquet column limit or contain columns without legacy
+   Parquet column IDs block the downgrade.
+
+### Preview the downgrade
+
+Use the `--dry-run` flag to list files that would be deleted without making
+any changes:
+
+```bash
+influxdb3 downgrade-to-parquet \
+  --cluster-id cluster01 \
+  --object-store file \
+  --data-dir ~/.influxdb3 \
+  --dry-run
+```
+
+### Run the downgrade
+
+```bash
+influxdb3 downgrade-to-parquet \
+  --cluster-id cluster01 \
+  --object-store file \
+  --data-dir ~/.influxdb3
+```
+
+After the downgrade completes, restart nodes without the `--use-pacha-tree` flag
+to resume standard Parquet storage mode.
+
+For all available options, see
+[Downgrade options](/influxdb3/enterprise/performance-preview/configure/#downgrade-options).
+
+## Export to Parquet
+
+You can export compacted data as Parquet files for use with external tools.
+
+> [!Note]
+> Data must be compacted before it can be exported.
+> Uncompacted data is not available for export at this time.
+
+### Export workflow
+
+```bash
+# Step 1: List available databases
+influxdb3 export databases
+
+# Step 2: List tables in a database
+influxdb3 export tables -d mydb
+
+# Step 3: List compacted 24-hour windows for a table
+influxdb3 export windows -d mydb -t cpu
+
+# Step 4: Export data as Parquet files
+influxdb3 export data -d mydb -t cpu -o ./export_output
+```
+
+To export specific time windows only:
+
+```bash
+influxdb3 export data -d mydb -t cpu -w 2026-01-15,2026-01-16 -o ./export_output
+```
+
+## Who should try the preview
+
+Consider enabling the preview in your staging or development environment if
+you have workloads with:
+
+- High cardinality or wide tables
+- Frequent backfill across time ranges
+- Query-heavy access patterns requiring low latency
+- Sparse schemas with dynamic column creation
+- Resource constraints where bounded memory and CPU usage matter
+
+> [!Important]
+> #### Important: New file format
+>
+> These upgrades use a new columnar file format (`.pt` files).
+> When you enable the preview, new data is written in the new format.
+> Hybrid query mode (enabled by default) allows querying across both legacy
+> Parquet data and new `.pt` data seamlessly.
+>
+> For the beta, we recommend starting with a fresh setup for
+> testing and evaluation rather than converting existing data.
+
+## Bug reports and feedback
+
+To share feedback on the performance upgrade preview:
+
+- Contact [InfluxData support](https://support.influxdata.com)
+- Reach out to your InfluxData account team
+
+Your feedback on stability and performance at scale helps shape the future of
+InfluxDB 3.
+
+{{< children hlevel="h2" readmore=true hr=true >}}
diff --git a/content/influxdb3/enterprise/performance-preview/configure.md b/content/influxdb3/enterprise/performance-preview/configure.md
new file mode 100644
index 0000000000..1b6a599c0c
--- /dev/null
+++ b/content/influxdb3/enterprise/performance-preview/configure.md
@@ -0,0 +1,467 @@
+---
+title: Configure the performance upgrade preview
+seotitle: Performance upgrade preview configuration reference for InfluxDB 3 Enterprise
+description: >
+  Complete reference for all configuration options available with the InfluxDB 3 Enterprise
+  performance upgrades, including WAL, snapshot, compaction, caching, and replication settings.
+menu:
+  influxdb3_enterprise:
+    name: Configuration reference
+    parent: Performance upgrade preview
+weight: 202
+influxdb3/enterprise/tags: [storage, configuration, beta, preview, reference]
+related:
+  - /influxdb3/enterprise/performance-preview/
+  - /influxdb3/enterprise/performance-preview/monitor/
+  - /influxdb3/enterprise/admin/performance-tuning/
+  - /influxdb3/enterprise/reference/config-options/
+---
+
+> [!Warning]
+> #### Performance preview beta
+> The performance upgrade preview is available to {{% product-name %}} Trial
+> and Commercial users as a beta. These features are subject to breaking changes
+> and **should not be used for production workloads**.
+
+This page provides a complete reference for all configuration options available
+with the performance upgrade preview.
+All `--pt-*` performance upgrade options require the `--use-pacha-tree` flag.
+
+If an option is omitted, the preview either derives a value from the existing
+`influxdb3 serve` configuration or falls back to an engine-specific default
+that balances resource usage and throughput.
+
+> [!Important]
+> Set `--num-io-threads` to the number of cores on the machine when using the
+> performance upgrade preview.
+
+- [General](#general)
+- [WAL](#wal)
+- [Snapshot](#snapshot)
+- [Gen0](#gen0)
+- [File cache](#file-cache)
+- [Replication (query nodes)](#replication-query-nodes)
+- [Compactor](#compactor)
+- [L1-L4 level tuning](#l1-l4-level-tuning)
+- [Example configurations](#example-configurations)
+- [Downgrade options](#downgrade-options)
+
+## General
+
+| Option | Description | Default |
+|:-------|:------------|:--------|
+| `--use-pacha-tree` | Enable the performance upgrade preview. Required for any other `--pt-` option to have effect. | Disabled |
+| `--pt-engine-path-prefix` | Optional path prefix for all engine data (WAL and compaction generations). Max 32 characters. Must start and end with alphanumeric; inner characters allow `[a-zA-Z0-9._-]`. Shorter paths improve partitioning in object stores. | No prefix |
+| `--pt-max-columns` | Maximum total columns across the entire instance. Must be at least 2. | `10,000,000` (10M) |
+| `--pt-enable-retention` | Enable retention enforcement. | `true` |
+| `--pt-disable-hybrid-query` | Disable hybrid query mode. When the preview is enabled with existing Parquet data, queries normally merge results across both Parquet and `.pt` files. Set this flag to query only `.pt` data. | `false` |
+| `--pt-enable-auto-dvc` | Enable automatic distinct value caching for `SHOW TAG VALUES` queries and the `tag_values()` SQL function. | Disabled |
+| `--pt-upgrade-poll-interval` | Polling interval for Parquet-to-PachaTree upgrade status monitoring. See [Upgrade from Parquet](/influxdb3/enterprise/performance-preview/#upgrade-from-parquet). | `5s` |
+
+### Engine path prefix
+
+Use a short prefix to improve partitioning in object stores:
+
+```bash
+influxdb3 serve \
+  # ...
+  --use-pacha-tree \
+  --pt-engine-path-prefix mydata
+```
+
+### Hybrid query mode
+
+When you enable the preview on an instance with existing Parquet data,
+hybrid query mode merges results across both Parquet and `.pt` files.
+Disable hybrid mode to query only `.pt` data:
+
+```bash
+influxdb3 serve \
+  # ...
+  --use-pacha-tree \
+  --pt-disable-hybrid-query
+```
+
+## WAL
+
+Configure Write-Ahead Log (WAL) behavior for durability and performance.
+
+| Option | Description | Default |
+|:-------|:------------|:--------|
+| `--pt-wal-flush-interval` | Flush interval for the WAL. | Inherits `--wal-flush-interval` (1s) |
+| `--pt-wal-flush-concurrency` | WAL flush concurrency. | `max(io_threads - 2, 2)` |
+| `--pt-wal-max-buffer-size` | Maximum in-memory WAL buffer before a flush is triggered regardless of the flush interval. Increase this if WAL files are flushed before the interval elapses. | `15MB` |
+| `--pt-wal-snapshots-to-keep` | Number of snapshot manifests worth of WAL history to retain. Must be greater than 0. | `5` |
+
+### WAL buffer size
+
+The WAL buffer accumulates incoming writes before flushing to object storage.
+Larger buffers reduce flush frequency and produce larger WAL files, but increase
+memory usage:
+
+```bash
+influxdb3 serve \
+  # ...
+  --use-pacha-tree \
+  --pt-wal-max-buffer-size 30MB
+```
+
+### Flush interval and concurrency
+
+Control how frequently the WAL flushes and how many workers run flushes in
+parallel:
+
+```bash
+influxdb3 serve \
+  # ...
+  --use-pacha-tree \
+  --pt-wal-flush-interval 2s \
+  --pt-wal-flush-concurrency 8
+```
+
+## Snapshot
+
+Configure snapshot buffer behavior, which controls how WAL files are merged
+into Gen0 files.
+
+| Option | Description | Default |
+|:-------|:------------|:--------|
+| `--pt-snapshot-size` | Maximum size of the active snapshot bucket before it is rotated for snapshotting. | `250MB` |
+| `--pt-snapshot-duration` | Time-based snapshot rotation trigger. Controls how often the ingester creates snapshots. Also used on query nodes as the bucket rotation interval for the replica buffer. | `10s` |
+| `--pt-max-concurrent-snapshots` | Maximum number of concurrent snapshot operations before applying backpressure to writers. | `5` |
+| `--pt-merge-threshold-size` | Maximum unmerged file size before triggering a merge operation. | `--pt-snapshot-size` / 4 (62.5MB) |
+
+### Snapshot size and duration
+
+Control when snapshot rotation triggers:
+
+```bash
+influxdb3 serve \
+  # ...
+  --use-pacha-tree \
+  --pt-snapshot-size 500MB \
+  --pt-snapshot-duration 15s
+```
+
+### Merge threshold
+
+Set the size threshold that triggers background merge operations.
+Lower values result in more frequent merges:
+
+```bash
+influxdb3 serve \
+  # ...
+  --use-pacha-tree \
+  --pt-merge-threshold-size 125MB
+```
+
+## Gen0
+
+Control the size of Gen0 files produced during merge operations.
+
+| Option | Description | Default |
+|:-------|:------------|:--------|
+| `--pt-gen0-max-rows-per-file` | Upper bound on rows per Gen0 file emitted during merge. | `10000000` (10M) |
+| `--pt-gen0-max-bytes-per-file` | Upper bound on bytes per Gen0 file emitted during merge. | `100MB` |
+
+### Gen0 file size limits
+
+Control the size of Gen0 files for query and compaction performance:
+
+```bash
+influxdb3 serve \
+  # ...
+  --use-pacha-tree \
+  --pt-gen0-max-rows-per-file 5000000 \
+  --pt-gen0-max-bytes-per-file 50MB
+```
+
+## File cache
+
+Configure data file caching for query performance.
+
+| Option | Description | Default |
+|:-------|:------------|:--------|
+| `--pt-file-cache-size` | Size of the data file cache (bytes or %). Set to `0` on dedicated ingest nodes. | Mirrors `--parquet-mem-cache-size` |
+| `--pt-disable-data-file-cache` | Disable data file caching. Set to `true` on dedicated ingest nodes. | `false` (automatically `true` if `--disable-parquet-mem-cache` is set) |
+| `--pt-file-cache-recency` | Only cache files newer than this age. Pre-caching on all-in-one and query nodes is based on this value. | Mirrors `--parquet-mem-cache-query-path-duration` (`3d`) |
+| `--pt-file-cache-evict-after` | Evict cached files that have not been read within this duration. | `24h` |
+
+> [!Note]
+> #### Dedicated ingest nodes
+> On dedicated ingest nodes (`--mode ingest`), disable the data file cache to avoid
+> wasting memory on data that ingest nodes never query.
+> Set `--pt-file-cache-size 0` or use `--pt-disable-data-file-cache`.
+> These options must be explicitly set—they are not applied automatically when
+> `--mode ingest` is used.
+> See [Disable caching on ingest nodes](#disable-caching-on-ingest-nodes) for an example.
+
+### File cache size
+
+Set the maximum size for the data file cache:
+
+```bash
+influxdb3 serve \
+  # ...
+  --use-pacha-tree \
+  --pt-file-cache-size 8GB
+```
+
+### Cache recency filter
+
+Only cache files containing data within a recent time window:
+
+```bash
+influxdb3 serve \
+  # ...
+  --use-pacha-tree \
+  --pt-file-cache-recency 24h
+```
+
+### Disable caching on ingest nodes
+
+For dedicated ingest nodes, disable the data file cache to save memory:
+
+```bash
+influxdb3 serve \
+  # ...
+  --use-pacha-tree \
+  --mode ingest \
+  --pt-disable-data-file-cache
+```
+
+## Replication (query nodes)
+
+Configure replication behavior for query nodes in distributed deployments.
+
+| Option | Description | Default |
+|:-------|:------------|:--------|
+| `--pt-wal-replication-interval` | Polling interval to check for new WAL files to replicate from ingest nodes. | `250ms` |
+| `--pt-wal-replica-recovery-concurrency` | Number of concurrent WAL file fetches during replica recovery or catchup. | `8` |
+| `--pt-wal-replica-steady-concurrency` | Number of concurrent WAL file fetches during steady-state replication. | `8` |
+| `--pt-wal-replica-queue-size` | Size of the queue between WAL file fetching and replica buffer merging. | `100` |
+| `--pt-wal-replica-recovery-tail-skip-limit` | Number of consecutive missing WAL files before stopping replica recovery. | `128` |
+| `--pt-replica-gen0-load-concurrency` | Limit on the number of Gen0 files loaded concurrently when the replica starts. | `16` |
+| `--pt-replica-max-buffer-size` | Maximum replica buffer size (bytes or %). Used by query nodes to store WAL files replicated from ingest nodes. | 50% of available memory, capped at 16GB |
+
+### Recovery concurrency
+
+Control parallelism during query node recovery or catchup:
+
+```bash
+influxdb3 serve \
+  # ...
+  --use-pacha-tree \
+  --mode query \
+  --pt-wal-replica-recovery-concurrency 16
+```
+
+### Steady-state replication
+
+Configure ongoing replication performance:
+
+```bash
+influxdb3 serve \
+  # ...
+  --use-pacha-tree \
+  --mode query \
+  --pt-wal-replica-steady-concurrency 4 \
+  --pt-wal-replica-queue-size 200
+```
+
+### Replica buffer size
+
+Control the maximum buffer size for replicated data on query nodes:
+
+```bash
+influxdb3 serve \
+  # ...
+  --use-pacha-tree \
+  --mode query \
+  --pt-replica-max-buffer-size 8GB
+```
+
+## Compactor
+
+Configure background compaction behavior.
+The compactor organizes data into fixed 24-hour UTC windows and progresses data
+through four compaction levels (L1 through L4).
+
+| Option | Description | Default |
+|:-------|:------------|:--------|
+| `--pt-partition-count` | Target number of partitions per compaction window. | `1` |
+| `--pt-compactor-input-size-budget` | Maximum total input bytes across all active compaction jobs. Acts as an admission control budget for the compactor scheduler. | 50% of system memory at startup |
+| `--pt-final-compaction-age` | Age threshold for final compaction. When all L1-L3 run sets in a window are older than this, a final compaction merges everything into L4. | `72h` |
+| `--pt-compactor-cleanup-cooldown` | Cooldown after checkpoint publish before replaced files can be cleaned up. | `10min` |
+
+### Compaction budget
+
+Control total memory allocated to active compaction jobs:
+
+```bash
+influxdb3 serve \
+  # ...
+  --use-pacha-tree \
+  --pt-compactor-input-size-budget 8GB
+```
+
+### Final compaction age
+
+Control when windows receive their final compaction into L4:
+
+```bash
+influxdb3 serve \
+  # ...
+  --use-pacha-tree \
+  --pt-final-compaction-age 48h
+```
+
+## L1-L4 level tuning
+
+These options control per-level compaction parameters.
+Data enters L1 from snapshot batch compaction and promotes through levels
+based on run set count triggers.
+
+| Level | Role | Default tail target | Default file size | Promotion trigger |
+|:------|:-----|:--------------------|:------------------|:------------------|
+| **L1** | Ingest landing zone | 600MB | 25MB | 3 run sets |
+| **L2** | First promotion tier | 1.2GB | 40MB | 3 run sets |
+| **L3** | Second promotion tier | 2.5GB | 75MB | 4 run sets |
+| **L4** | Terminal (fully compacted) | 50GB | 125MB | N/A |
+
+### L1 options
+
+| Option | Description | Default |
+|:-------|:------------|:--------|
+| `--pt-l1-tail-target-bytes` | L1 tail run set target size. | `600MB` |
+| `--pt-l1-target-file-bytes` | L1 target file size. | `25MB` |
+| `--pt-l1-promotion-count` | Number of L1 run sets that triggers promotion to L2. | `3` |
+
+### L2 options
+
+| Option | Description | Default |
+|:-------|:------------|:--------|
+| `--pt-l2-tail-target-bytes` | L2 tail run set target size. | `1.2GB` |
+| `--pt-l2-target-file-bytes` | L2 target file size. | `40MB` |
+| `--pt-l2-promotion-count` | Number of L2 run sets that triggers promotion to L3. | `3` |
+
+### L3 options
+
+| Option | Description | Default |
+|:-------|:------------|:--------|
+| `--pt-l3-tail-target-bytes` | L3 tail run set target size. | `2.5GB` |
+| `--pt-l3-target-file-bytes` | L3 target file size. | `75MB` |
+| `--pt-l3-promotion-count` | Number of L3 run sets that triggers promotion to L4. | `4` |
+
+### L4 options
+
+| Option | Description | Default |
+|:-------|:------------|:--------|
+| `--pt-l4-tail-target-bytes` | L4 tail run set target size. | `50GB` |
+| `--pt-l4-target-file-bytes` | L4 target file size. | `125MB` |
+
+## Example configurations
+
+### Development (minimal resources)
+
+```bash
+influxdb3 serve \
+  --node-id dev01 \
+  --cluster-id dev-cluster \
+  --object-store file \
+  --data-dir ~/.influxdb3 \
+  --use-pacha-tree \
+  --num-io-threads 2 \
+  --pt-file-cache-size 512MB \
+  --pt-wal-max-buffer-size 5MB \
+  --pt-snapshot-size 100MB
+```
+
+### Production all-in-one (8 cores, 32 GB RAM)
+
+```bash
+influxdb3 serve \
+  --node-id prod01 \
+  --cluster-id prod-cluster \
+  --object-store s3 \
+  --bucket S3_BUCKET \
+  --aws-access-key-id AWS_ACCESS_KEY_ID \
+  --aws-secret-access-key AWS_SECRET_ACCESS_KEY \
+  --use-pacha-tree \
+  --num-io-threads 8 \
+  --pt-file-cache-size 8GB \
+  --pt-wal-max-buffer-size 30MB \
+  --pt-snapshot-size 500MB \
+  --pt-wal-flush-concurrency 4
+```
+
+### High-throughput ingest node
+
+```bash
+influxdb3 serve \
+  --node-id ingest01 \
+  --cluster-id prod-cluster \
+  --object-store s3 \
+  --bucket S3_BUCKET \
+  --aws-access-key-id AWS_ACCESS_KEY_ID \
+  --aws-secret-access-key AWS_SECRET_ACCESS_KEY \
+  --use-pacha-tree \
+  --mode ingest \
+  --num-io-threads 16 \
+  --pt-wal-max-buffer-size 50MB \
+  --pt-wal-flush-interval 2s \
+  --pt-wal-flush-concurrency 8 \
+  --pt-snapshot-size 1GB \
+  --pt-disable-data-file-cache
+```
+
+### Query-optimized node
+
+```bash
+influxdb3 serve \
+  --node-id query01 \
+  --cluster-id prod-cluster \
+  --object-store s3 \
+  --bucket S3_BUCKET \
+  --aws-access-key-id AWS_ACCESS_KEY_ID \
+  --aws-secret-access-key AWS_SECRET_ACCESS_KEY \
+  --use-pacha-tree \
+  --mode query \
+  --num-io-threads 16 \
+  --pt-file-cache-size 16GB \
+  --pt-file-cache-recency 24h \
+  --pt-replica-max-buffer-size 8GB
+```
+
+### Dedicated compactor
+
+```bash
+influxdb3 serve \
+  --node-id compact01 \
+  --cluster-id prod-cluster \
+  --object-store s3 \
+  --bucket S3_BUCKET \
+  --aws-access-key-id AWS_ACCESS_KEY_ID \
+  --aws-secret-access-key AWS_SECRET_ACCESS_KEY \
+  --use-pacha-tree \
+  --mode compact \
+  --num-io-threads 8 \
+  --pt-compactor-input-size-budget 12GB
+```
+
+## Downgrade options
+
+The `influxdb3 downgrade-to-parquet` command reverts a cluster from the
+performance preview back to standard Parquet storage.
+For the downgrade procedure, see
+[Downgrade to Parquet](/influxdb3/enterprise/performance-preview/#downgrade-to-parquet).
+
+| Option | Description |
+|:-------|:------------|
+| `--cluster-id` | _(Required)_ Cluster identifier. |
+| `--object-store` | _(Required)_ Object storage type (`file`, `s3`, `gcs`, `azure`). |
+| `--data-dir` | Location of data files for a local (`file`) object store. |
+| `--bucket` | Object store bucket name (for `s3`, `gcs`, `azure`). |
+| `--dry-run` | Preview mode--list files that would be deleted without making changes. |
+| `--yes` | Skip the confirmation prompt. |
+| `--ignore-running` | Proceed even if nodes appear to be running. **Warning:** may cause data inconsistency if nodes are actively writing. |
diff --git a/content/influxdb3/enterprise/performance-preview/monitor.md b/content/influxdb3/enterprise/performance-preview/monitor.md
new file mode 100644
index 0000000000..958c2e745f
--- /dev/null
+++ b/content/influxdb3/enterprise/performance-preview/monitor.md
@@ -0,0 +1,354 @@
+---
+title: Monitor the performance upgrade preview
+seotitle: Monitor the performance upgrade preview in InfluxDB 3 Enterprise
+description: >
+  Use system tables and query telemetry to monitor file status, query execution,
+  and overall performance when using InfluxDB 3 Enterprise performance upgrades.
+menu:
+  influxdb3_enterprise:
+    name: Monitor
+    parent: Performance upgrade preview
+weight: 203
+influxdb3/enterprise/tags: [storage, monitoring, beta, preview, system tables]
+related:
+  - /influxdb3/enterprise/performance-preview/
+  - /influxdb3/enterprise/performance-preview/configure/
+  - /influxdb3/enterprise/admin/query-system-data/
+---
+
+> [!Warning]
+> #### Performance preview beta
+> The performance upgrade preview is available to {{% product-name %}} Trial
+> and Commercial users as a beta. These features are subject to breaking changes
+> and **should not be used for production workloads**.
+
+{{% product-name %}} provides system tables and a query telemetry endpoint to
+monitor file status, query execution, and overall performance when using the
+performance upgrade preview.
+
+## System tables
+
+The upgraded storage engine exposes internal state through system tables that
+you can query with SQL.
+
+### system.pt_ingest_wal
+
+View WAL files and their partitions:
+
+```sql
+SELECT * FROM system.pt_ingest_wal;
+```
+
+Example output:
+
+| wal_file_id | partition_id | database_id | table_id | min_time | max_time | row_count | size_bytes |
+|:------------|:-------------|:------------|:---------|:---------|:---------|:----------|:-----------|
+| wal_001 | p_1 | db_1 | t_1 | 2024-01-01T00:00:00Z | 2024-01-01T00:10:00Z | 50000 | 2456789 |
+| wal_002 | p_1 | db_1 | t_1 | 2024-01-01T00:10:00Z | 2024-01-01T00:20:00Z | 48000 | 2345678 |
+
+Use this table to monitor:
+
+- **WAL accumulation**: Track the number and size of unmerged WAL files
+- **Partition distribution**: See how data is distributed across partitions
+- **Time coverage**: Verify data time ranges
+
+#### Monitor WAL backlog
+
+Check for WAL accumulation that may indicate merging is falling behind:
+
+```sql
+SELECT
+  COUNT(*) as wal_file_count,
+  SUM(size_bytes) / 1024 / 1024 as total_size_mb,
+  MIN(min_time) as oldest_data,
+  MAX(max_time) as newest_data
+FROM system.pt_ingest_wal;
+```
+
+### system.pt_ingest_files
+
+View Gen0 files with metadata:
+
+```sql
+SELECT * FROM system.pt_ingest_files;
+```
+
+Example output:
+
+| file_id | generation | database_id | table_id | min_time | max_time | row_count | size_bytes |
+|:--------|:-----------|:------------|:---------|:---------|:---------|:----------|:-----------|
+| gen0_001 | 0 | db_1 | t_1 | 2024-01-01T00:00:00Z | 2024-01-01T01:00:00Z | 500000 | 45678901 |
+| gen0_002 | 0 | db_1 | t_1 | 2024-01-01T01:00:00Z | 2024-01-01T02:00:00Z | 480000 | 43567890 |
+
+Use this table to monitor:
+
+- **File counts per generation**: Track compaction progress
+- **File sizes**: Verify files are within configured limits
+- **Time ranges**: Identify Gen0 files that span multiple compaction windows
+
+#### Monitor file distribution
+
+Check file distribution and compaction status:
+
+```sql
+SELECT
+  generation,
+  COUNT(*) as file_count,
+  SUM(row_count) as total_rows,
+  SUM(size_bytes) / 1024 / 1024 as total_size_mb,
+  AVG(size_bytes) / 1024 / 1024 as avg_file_size_mb
+FROM system.pt_ingest_files
+GROUP BY generation
+ORDER BY generation;
+```
+
+## Parquet upgrade status
+
+If you [upgraded from Parquet](/influxdb3/enterprise/performance-preview/#upgrade-from-parquet),
+use these system tables to monitor migration progress.
+
+### system.upgrade_parquet_node
+
+View per-node upgrade status:
+
+```sql
+SELECT * FROM system.upgrade_parquet_node;
+```
+
+Monitor this table to confirm each node reaches `completed` status.
+During the upgrade, nodes progress through detection, conversion, and
+finalization stages.
+
+### system.upgrade_parquet
+
+View per-file migration progress:
+
+```sql
+SELECT * FROM system.upgrade_parquet;
+```
+
+Use this table to track individual file conversions during the migration.
+The status updates on a polling interval (default 5 seconds, configurable with
+`--pt-upgrade-poll-interval`).
+
+## Query telemetry
+
+The query telemetry endpoint provides detailed execution statistics for
+analyzing query performance.
+
+### Enable query telemetry
+
+Query the telemetry endpoint after executing a query:
+
+```bash
+curl -X GET "http://localhost:8181/api/v3/query_sql_telemetry" \
+  -H "Authorization: Bearer AUTH_TOKEN"
+```
+
+Replace `AUTH_TOKEN` with your authentication token.
+
+### Telemetry response
+
+The response includes:
+
+| Field | Description |
+|:------|:------------|
+| `query_id` | Unique identifier for the query |
+| `execution_time_us` | Total execution time in microseconds |
+| `chunks` | Per-chunk statistics |
+| `cache_stats` | Cache hit rates by type |
+| `file_stats` | File-level read statistics |
+
+### Example telemetry output
+
+```json
+{
+  "query_id": "q_12345",
+  "execution_time_us": 4523,
+  "chunks": [
+    {
+      "chunk_id": "c_1",
+      "files_scanned": 3,
+      "blocks_processed": 12,
+      "rows_read": 24000,
+      "rows_returned": 150,
+      "bytes_read": 1234567
+    }
+  ],
+  "cache_stats": {
+    "gen0_hits": 5,
+    "gen0_misses": 1,
+    "compacted_hits": 8,
+    "compacted_misses": 2
+  }
+}
+```
+
+## Performance analysis
+
+### Query performance metrics
+
+Track these key metrics for query performance:
+
+| Metric | Good | Warning | Action |
+|:-------|:-----|:--------|:-------|
+| Cache hit rate | >80% | <60% | Increase `--pt-file-cache-size` or `--pt-file-cache-recency` |
+| Rows read vs returned ratio | <100:1 | >1000:1 | Add more selective predicates |
+
+### Ingest performance metrics
+
+Monitor these metrics for write performance:
+
+| Metric | Healthy | Warning | Action |
+|:-------|:--------|:--------|:-------|
+| WAL file count | <50 | >100 | Increase `--pt-wal-flush-concurrency` |
+| Gen0 file count | <100 | >200 | Increase `--pt-compactor-input-size-budget` |
+
+### Monitor with SQL
+
+Create a performance summary query:
+
+```sql
+-- File generation summary
+SELECT
+  'Gen0 files' as metric,
+  COUNT(*) as count,
+  SUM(size_bytes) / 1024 / 1024 as size_mb
+FROM system.pt_ingest_files
+WHERE generation = 0
+
+UNION ALL
+
+SELECT
+  'Compacted files' as metric,
+  COUNT(*) as count,
+  SUM(size_bytes) / 1024 / 1024 as size_mb
+FROM system.pt_ingest_files
+WHERE generation > 0
+
+UNION ALL
+
+SELECT
+  'WAL files' as metric,
+  COUNT(*) as count,
+  SUM(size_bytes) / 1024 / 1024 as size_mb
+FROM system.pt_ingest_wal;
+```
+
+## Troubleshooting
+
+### High WAL file count
+
+**Symptom**: `system.pt_ingest_wal` shows many accumulated files.
+
+**Possible causes**:
+
+- Merge operations falling behind write rate
+- Insufficient flush concurrency
+- Object storage latency
+
+**Solutions**:
+
+1. Increase flush concurrency:
+   ```bash
+   --pt-wal-flush-concurrency 8
+   ```
+
+2. Increase WAL flush interval to create larger, fewer files:
+   ```bash
+   --pt-wal-flush-interval 5s
+   ```
+
+3. Increase the WAL buffer size so each flush produces a larger file:
+   ```bash
+   --pt-wal-max-buffer-size 30MB
+   ```
+
+4. Check object storage performance and connectivity.
+
+### High cache miss rate
+
+**Symptom**: `cache_stats` shows >40% miss rate.
+
+**Possible causes**:
+
+- Cache size too small for working set
+- Cache recency window too narrow
+- Random access patterns across time ranges
+
+**Solutions**:
+
+1. Increase cache size:
+   ```bash
+   --pt-file-cache-size 16GB
+   ```
+
+2. Extend cache recency window:
+   ```bash
+   --pt-file-cache-recency 24h
+   ```
+
+3. Extend eviction timeout:
+   ```bash
+   --pt-file-cache-evict-after 48h
+   ```
+
+### Slow compaction
+
+**Symptom**: Gen0 file count continues to grow.
+
+**Possible causes**:
+
+- Compaction budget too low for write volume
+- High write rate overwhelming compaction
+- Snapshot size too large, creating oversized Gen0 files
+
+**Solutions**:
+
+1. Increase the compaction input size budget:
+   ```bash
+   --pt-compactor-input-size-budget 12GB
+   ```
+
+2. Reduce snapshot size to create smaller, more frequent Gen0 files:
+   ```bash
+   --pt-snapshot-size 125MB
+   ```
+
+3. For distributed deployments, add a dedicated compactor node:
+   ```bash
+   influxdb3 serve \
+     # ...
+     --use-pacha-tree \
+     --mode compact
+   ```
+
+### Query node lag
+
+**Symptom**: Query nodes return stale data.
+
+**Possible causes**:
+
+- Replication falling behind
+- Network latency to object storage
+- Insufficient replica concurrency
+
+**Solutions**:
+
+For a full list of replication options, see
+[Replication (query nodes)](/influxdb3/enterprise/performance-preview/configure/#replication-query-nodes).
+
+1. Increase replication concurrency:
+   ```bash
+   --pt-wal-replica-steady-concurrency 8
+   ```
+
+2. Reduce the replication polling interval:
+   ```bash
+   --pt-wal-replication-interval 100ms
+   ```
+
+3. Increase replica queue size:
+   ```bash
+   --pt-wal-replica-queue-size 200
+   ```