Memgraph in production guides (general and GraphRAG) #1239
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
The latest updates on your projects. Learn more about Vercel for Git ↗︎
|
Josipmrden
changed the title
matea16
reviewed
Apr 24, 2025
| * [functions](/querying/functions): return a null value | ||
|
|
||
| Please note that deleting same part of the graph from parallel transaction will lead to undefined behavior. | ||
| **Please note that deleting same part of the graph from parallel transaction will lead to undefined behavior.** |
Contributor
There was a problem hiding this comment.
Suggested change
| **Please note that deleting same part of the graph from parallel transaction will lead to undefined behavior.** | |
| <Callout type="warning"> | |
| Please note that deleting the same part of the graph from a parallel transaction will lead to undefined behavior. | |
| </Callout> |
| | Cores | 1 vCPU | ≥ 8 vCPUs (≥ 4 physical cores) | | ||
| | Network | 100 Mbps | ≥ 1 Gbps | | ||
|
|
||
| The disk is used for storing database [durability files](/configuration/data-durability-and-backup) - snapshots and write-ahead |
Contributor
There was a problem hiding this comment.
Suggested change
| The disk is used for storing database [durability files](/configuration/data-durability-and-backup) - snapshots and write-ahead | |
| The disk is used for storing database [durability files](/configuration/data-durability-and-backup), including snapshots and write-ahead |
| | Network | 100 Mbps | ≥ 1 Gbps | | ||
|
|
||
| The disk is used for storing database [durability files](/configuration/data-durability-and-backup) - snapshots and write-ahead | ||
| logs. By default, Memgraph stores 3 latest snapshots in the database (flag to adjust this is `--storage-snapshot-retention-count`). |
Contributor
There was a problem hiding this comment.
Suggested change
| logs. By default, Memgraph stores 3 latest snapshots in the database (flag to adjust this is `--storage-snapshot-retention-count`). | |
| logs. By default, Memgraph stores the **three most recent snapshots** in the database (this can be configured using the `--storage-snapshot-retention-count` flag). | |
Comment on lines
124
to
125
| The amount of CPU cores varies per use case. For horizontal scalability, you can further increase the amount of cores on your | ||
| system for additional scalability. |
Contributor
There was a problem hiding this comment.
Suggested change
| The amount of CPU cores varies per use case. For horizontal scalability, you can further increase the amount of cores on your | |
| system for additional scalability. | |
| The number of CPU cores required depends on your specific use case. For horizontal scalability, you can increase the number of available cores on your system for additional scalability. |
pages/memgraph-in-production.mdx
Outdated
Comment on lines
11
to
14
| When deploying Memgraph in production, it is essential to consider a set of prerequisites to ensure optimal performance, | ||
| scalability, and resilience. This includes hardware considerations, the correct sizing of instances, configuring drivers, using | ||
| appropriate flags when starting Memgraph, importing data, and connecting to external sources. The guidelines in this section will | ||
| help you make informed decisions for your specific use case, ensuring that Memgraph performs effectively in your environment. |
Contributor
There was a problem hiding this comment.
Suggested change
| When deploying Memgraph in production, it is essential to consider a set of prerequisites to ensure optimal performance, | |
| scalability, and resilience. This includes hardware considerations, the correct sizing of instances, configuring drivers, using | |
| appropriate flags when starting Memgraph, importing data, and connecting to external sources. The guidelines in this section will | |
| help you make informed decisions for your specific use case, ensuring that Memgraph performs effectively in your environment. | |
| When deploying Memgraph in production, it is essential to consider a set of prerequisites to ensure optimal **performance**, **scalability** and **resilience**. That includes decisions about hardware configurations and integration strategies. | |
| This guide is your starting point to production-readiness with Memgraph. | |
| ## ✅ What you'll need to consider | |
| Before you dive into specific setups, it's important to think about: | |
| - **Hardware requirements** and **instance sizing** | |
| - **Driver configuration** | |
| - **Flags** when starting Memgraph | |
| - **Data import** best practices | |
| - Connecting to **external sources** | |
| These factors ensure Memgraph performs effectively in your environment, no matter the use case. |
|
|
||
| Reason for that is because environment variables always override any system settings that are set via queries. | ||
|
|
||
| Additionally, please check out [how to set up the Memgraph Lab Enterprise license](/data-visualization/user-manual/remote-storage#how-to-set-it-up) |
Contributor
There was a problem hiding this comment.
Suggested change
| Additionally, please check out [how to set up the Memgraph Lab Enterprise license](/data-visualization/user-manual/remote-storage#how-to-set-it-up) | |
| Additionally, please check out [how to set up the Memgraph Lab Enterprise license](/memgraph-lab/configuration#adding-memgraph-enterprise-license) | |
| Memgraph fully supports the **Cypher query language**, making it easy to express complex graph patterns. In addition to Cypher, | ||
| Memgraph has **built-in path traversal capabilities** at the core of the database, enabling **lightning-fast traversals** optimized for | ||
| performance-critical use cases. You can learn more about these in our | ||
| [Deep Path Traversal guide](/advanced-algorithms/deep-path-traversal). |
Contributor
There was a problem hiding this comment.
Suggested change
| [Deep Path Traversal guide](/advanced-algorithms/deep-path-traversal). | |
| [Deep path traversal guide](/advanced-algorithms/deep-path-traversal). | |
| **[MAGE library](/advanced-algorithms/available-algorithms)**, which includes a wide range of pre-built | ||
| **graph algorithms** and **procedures** for tasks like community detection, centrality scoring, node similarity, and more. | ||
|
|
||
| We encourage users to explore the other guides in the **"Memgraph in Production"** series, where you'll find detailed examples and |
Contributor
There was a problem hiding this comment.
Suggested change
| We encourage users to explore the other guides in the **"Memgraph in Production"** series, where you'll find detailed examples and | |
| We encourage users to explore the other guides in the *Memgraph in Production* series, where you'll find detailed examples and | |
| description: General suggestions when working with Memgraph, from testing to production. | ||
| --- | ||
|
|
||
| import { Callout } from 'nextra/components' |
Contributor
There was a problem hiding this comment.
Suggested change
| import { Callout } from 'nextra/components' | |
| import { Callout } from 'nextra/components' | |
| import {CommunityLinks} from '/components/social-card/CommunityLinks' | |
| import { Steps } from 'nextra/components' |
| **graph algorithms** and **procedures** for tasks like community detection, centrality scoring, node similarity, and more. | ||
|
|
||
| We encourage users to explore the other guides in the **"Memgraph in Production"** series, where you'll find detailed examples and | ||
| recommendations on which types of queries are most effective based on your specific **workload and use case**. |
Contributor
There was a problem hiding this comment.
Suggested change
| recommendations on which types of queries are most effective based on your specific **workload and use case**. | |
| recommendations on which types of queries are most effective based on your specific **workload and use case**. | |
| <CommunityLinks/> |
matea16
requested changes
Apr 25, 2025
| page. It provides **foundational, use-case-agnostic advice** for deploying Memgraph in production. | ||
|
|
||
| This guide builds on that foundation, offering **additional recommendations tailored to specific workloads**. | ||
| In cases where guidance overlaps, the information in this chapter should be seen as **complementary or overriding**, depending |
Contributor
There was a problem hiding this comment.
Suggested change
| In cases where guidance overlaps, the information in this chapter should be seen as **complementary or overriding**, depending | |
| In cases where guidance overlaps, consider the information here as **complementary or overriding**, depending | |
| on the unique needs of your use case. | ||
| </Callout> | ||
|
|
||
| ## When to use this guide |
Contributor
There was a problem hiding this comment.
Suggested change
| ## When to use this guide | |
| ## Is this guide for you? | |
| ## When to use this guide | ||
|
|
||
| This guide is for you if you're exploring or building **GraphRAG (Graph-Augmented Retrieval-Augmented Generation)** systems. | ||
| Consider diving into this guide when: |
Contributor
There was a problem hiding this comment.
Suggested change
| Consider diving into this guide when: | |
| You'll benefit from this content if: | |
Comment on lines
27
to
28
| - 🔄 You need to **seamlessly extract knowledge graphs from multiple source systems**, especially when graph representation | ||
| naturally suits the data structure. |
Contributor
There was a problem hiding this comment.
Suggested change
| - 🔄 You need to **seamlessly extract knowledge graphs from multiple source systems**, especially when graph representation | |
| naturally suits the data structure. | |
| - 🔄 You need to **seamlessly extract knowledge graphs from multiple source systems**, especially when a graph structure naturally fits the data. | |
| If any of these resonate with your project, this guide will walk you through the best practices and configurations to bring | ||
| GraphRAG to life using Memgraph. | ||
|
|
||
| ## Why should you choose Memgraph for GraphRAG use cases |
Contributor
There was a problem hiding this comment.
Suggested change
| ## Why should you choose Memgraph for GraphRAG use cases | |
| ## Why choose Memgraph for GraphRAG use cases? | |
| Understand which **enterprise features** — such as security, access controls, and dynamic graph algorithms are | ||
| essential for production-ready GraphRAG deployments. | ||
|
|
||
| - **[Queries that best suit your workload](#queries-that-best-suit-your-workload)** |
Contributor
There was a problem hiding this comment.
Suggested change
| - **[Queries that best suit your workload](#queries-that-best-suit-your-workload)** | |
| - [Queries that best suit your workload](#queries-that-best-suit-your-workload) | |
| Learn how to use **deep path traversals**, **vector search**, and **dynamic MAGE algorithms** to efficiently retrieve contextual data and | ||
| handle **high-velocity graphs**. | ||
|
|
||
| - **[Memgraph ecosystem](#memgraph-ecosystem)** |
Contributor
There was a problem hiding this comment.
Suggested change
| - **[Memgraph ecosystem](#memgraph-ecosystem)** | |
| - [Memgraph ecosystem](#memgraph-ecosystem) | |
| - 🧭 **Reference-based indexing**: Embeddings will be stored only in the vector index, with the property storage holding just a reference. | ||
| Since embeddings are used exclusively for vector search, this eliminates duplication. | ||
|
|
||
| - ⚙️ **Support for float16 in usearch**: Users will be able to store embeddings as 2-byte floats (float16), commonly used in neural networks. |
Contributor
There was a problem hiding this comment.
Suggested change
| - ⚙️ **Support for float16 in usearch**: Users will be able to store embeddings as 2-byte floats (float16), commonly used in neural networks. | |
| - ⚙️ **Support for float16 in usearch**: Users will be able to store embeddings as 2-byte floats (`float16`), commonly used in neural networks. | |
Comment on lines
110
to
118
| - Consider the **appropriate embedding dimension** for your use case. While models like OpenAI use 1536 or 3072 dimensions, **lower-dimensional vectors | ||
| (e.g., 512 or 768)** often result in only a **5–6% drop in accuracy** and drastically reduce memory consumption. | ||
|
|
||
| - If in-memory embedding storage is too demanding, you can **offload embeddings to a third-party vector database** and still integrate it | ||
| into your GraphRAG pipeline alongside Memgraph. | ||
|
|
||
| - If you're also storing **document context** in Memgraph, keep in mind that these long strings are currently stored in memory as well. A short-term | ||
| roadmap item will enable **offloading static text content to disk**, since these strings rarely change. This will further | ||
| **increase your memory efficiency and scalability**. |
Contributor
There was a problem hiding this comment.
Suggested change
| - Consider the **appropriate embedding dimension** for your use case. While models like OpenAI use 1536 or 3072 dimensions, **lower-dimensional vectors | |
| (e.g., 512 or 768)** often result in only a **5–6% drop in accuracy** and drastically reduce memory consumption. | |
| - If in-memory embedding storage is too demanding, you can **offload embeddings to a third-party vector database** and still integrate it | |
| into your GraphRAG pipeline alongside Memgraph. | |
| - If you're also storing **document context** in Memgraph, keep in mind that these long strings are currently stored in memory as well. A short-term | |
| roadmap item will enable **offloading static text content to disk**, since these strings rarely change. This will further | |
| **increase your memory efficiency and scalability**. | |
| - **Choose the right embedding dimension**: | |
| While models like OpenAI use 1536 or 3072 dimensions, **lower-dimensional vectors like 512 or 768** often result in only a **5–6% drop in accuracy** and drastically reduce memory consumption. | |
| - **Offload if needed**: If in-memory embedding storage is too demanding, you can **offload embeddings to a third-party vector database** and still integrate it | |
| into your GraphRAG pipeline alongside Memgraph. | |
| - **Watch for string context size**: If you're also storing **document context** in Memgraph, keep in mind that these long strings are currently stored in memory as well. A short-term | |
| roadmap item will enable **offloading static text content to disk**, since these strings rarely change. This will further | |
| **increase your memory efficiency and scalability**. | |
|
|
||
|
|
||
| ## Queries that best suit your workload | ||
| When integrating Memgraph into open-source GraphRAG frameworks like **LlamaIndex** or **LangGraph**, |
Contributor
There was a problem hiding this comment.
Suggested change
| When integrating Memgraph into open-source GraphRAG frameworks like **LlamaIndex** or **LangGraph**, | |
| When integrating Memgraph into open-source GraphRAG frameworks like [LlamaIndex](/ai-ecosystem/integrations#llamaindex), [LangGraph](/ai-ecosystem/integrations#langchain) or [MCP](/ai-ecosystem/integrations#model-context-protocol-mcp), | |
matea16
approved these changes
Apr 25, 2025
* Update initial page * Add page for evaluating memgraph -> mgbench * Add title for evaluating memgraph * Address PR comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Release note
General production guide, and GraphRAG production guide
Related product PRs
PRs from product repo this doc page is related to:
(paste the links to the PRs)
Checklist:
bugfixorfeaturelabel, based on the product PR type you're documenting