Run benchmarks [run-benchmarks schemas,messages,stateful-storage]

designdocs/schema_v2.md

@@ -0,0 +1,41 @@
+# On Chain Message Storage
+
+## Context and Scope
+The proposed feature consists of changes that is going to be one (or more) pallet(s) in runtime of a
+Substrate based blockchain, and it will be used in all environments including production.
+
+## Problem Statement
+After introduction of **Proof of Validity** or **PoV** in runtime weights, all pallets should be
+re-evaluated and refactored if necessary to minimize the usage of **PoV**. This is to ensure all
+important operations are scalable.
+This document tries to propose some changes on **Schemas** pallet to optimize the **PoV** size.
+
+## Goals
+- Minimizing Weights including **execution times** and **PoV** size.
+
+## Proposal
+Split Schemas into `SchemaInfo` and `payload` would allow lower **PoV** when verifying schema existence
+or compatibility.
+
+### Main Storage types
+- **SchemaInfos**
+    - _Type_: `StorageMap<SchemaId, SchemaInfo>`
+    - _Purpose_: Main structure To store related properties of any schema
+      index
+- **SchemaPayloads**
+    -  _Type_: `StorageMap<SchemaId, BoundedVec<u8>>`
+    - _Purpose_: Stores the payload or model for each schema
+
+
+### On Chain Structure
+Following is a proposed data structure for storing schema information on chain.
+```rust
+pub struct SchemaInfo {
+    /// The type of model (AvroBinary, Parquet, etc.)
+    pub model_type: ModelType,
+    /// The payload location
+    pub payload_location: PayloadLocation,
+    /// additional control settings for the schema
+    pub settings: SchemaSettings,
+}
+```