Notes from the Trenches

Graphistry has made it to Azure! The easiest way to get started in Azure is through the Azure Marketplace, and for enterprise users, we also support Docker installation.

Of course, we still love Amazon too: The Graphistry ontology now supports CloudTrails datatypes out-of-the-box. On-premise? In 2.24.11, we also added an example of setting up an RHEL 7.6 environment for nvidia-docker-based software in addition to our existing Ubuntu guides.  

Version 2.23.4 brings a bunch to enterprise analysts and developers! Read on to learn more about those, and see full release notes at our new release notes page . 

Version 2.22.7 streamlines use with Gremlin/TinkerPop (CosmosDB, Neptune, JanusGraph, ...), helps you turn Jupyter Notebooks into dashboards, and adds more URL API parameters. Read on to learn more about those, and see full release notes at our new release notes page . 

 

Curious to turn your graph DB data into GPU-accelerated visual graph analytics sessions & dashboards? One-click launch on AWS now! 

 

 

 

Version 2.22 makes life better for both new and existing users. We're especially excited about introducing TigerGraph and SQL support, and the continued progress with the 2.0 engine. Read on to learn more about those, and see full release notes at our new release notes page . 

 

Curious to explore & automate your own CSV or database/API data with GPU visual graph analytics and investigation automation? One-click launch on AWS now! 

 

 

 

 

 

It feels likes eye-popping times for those deep into building the future of visual data experiences. With Looker exiting (-> Google for $3B), Tableau exiting (->SalesForce for $16B), and less public, Periscope & ZoomData exiting, the Graphistry team is experiencing good feelings and key reflections. One of them is... the $16B exits are just a prelude to the next $160B in opportunities.

 

One of the easiest and most powerful ways to empower your team is to create and embed automated Investigation Templates (docs). Analysts don't need to know what templates are available ahead of time: instead, they get Graphistry links embedded into their existing workflows. For example, you can augment alert emails with targeted investigation links, or add contextual links to any web dashboard. This is great for tasks like recommending particular kinds of investigations, and putting contextual entity views in reach at the right time.

 

The video tutorial walks through creating an investigation template and embedding links into Splunk as contextual Workflow Actions:

 

 

 

 

Next steps & further reading

• Docs for investigation template links
• Graphistry 2.0  + Launch in your AWS

 

Graphistry makes it easy to explore the hidden connections in any CSV or flat file by automatically exposing the underlying graph. This tutorial walks through the CSV Mini-App notebook that comes with Graphistry and applies it to visualizing the recent Implant Files medical device recalls database by the ICIJ.

 

Screenshot: ICIJ's The Implant Files visualized live with Graphistry - The pandemic of 70,000+ medical device recalls

 

1. Setup

• Download the recall data from the ICIJ and unzip to get the event data, events-1551346702.csv 
• Launch Graphistry!

 

2. Go through the video tutorial!

 

 

• Launch and clone the CSV Upload Mini-App notebook, and rename to "icij_implants.ipynb"
• Follow the instructions in the notebook
• Settings used for each section:
  
  • Upload:
    
    file_path = './events-1551346702.csv'
    
    
  • Data cleaning:
    
    hits = pd.DataFrame([[c, len(df[c].unique())] for c in df.columns], columns=['col', 'num_uniq']).sort_values('num_uniq')
    
    skip_nodes = ['icij_notes', 'determined_cause', 'action_classification', 'icij_notes', 'country', 'status', 'source']
    nodes = [x for x in list(hits.query('num_uniq > 10 & num_uniq < 9288')['col']) if not x in skip_nodes]
    
    df = df_orig.query('country == "USA"')
    
    
  • Plotting:
    
    mode = 'B'
    max_rows = 50000
    node_cols = nodes
    categories = { }
    
    

Next steps & further reading

• Pandas pd.read_csv() : Load most files
• Graphistry's graphistry.hypergraph(): Turn any Pandas dataframe into a live graph visualization
• Graphistry in Jupyter notebooks + UI guide 
• ICIJ's Implant Files investigation (+ the data!)
• Resulting use of CSV Mini-App notebook
• Graphistry 2.0  + Launch in your AWS

 

Last Updated: 9/7/2019

 

AWS Marketplace makes it easy to launch Graphistry. This tutorial shares a visual guide of launching Graphistry, trying your first notebook, and help with common configuration options and occasional error handling.  When you're ready, head to AWS Marketplace! And, if you haven't heard about Graphistry's  AWS launch, learn about why we did it and what's new.

 

Launch a private Graphistry instance from AWS Marketplace

 

 

  

1. Subscribe Screen

• If prompted, accept terms and wait through "Pending" period
  • Do not configure the annual contract discounted prepay option: Total contract price should be $0
• Continue to Configuration

 

2. Configure Screen

• Version: Default - the latest Graphistry version
• Region: Pick a GPU-capable region close to your users or data
  • Tip: Search for P3 series (Nvidia V100)  on the EC2 instance comparison chart, with US East and Oregon as great options in the US.
  • Do not configure the annual contract discounted prepay option: Total contract price should be $0
• Continue to Launch

 

3. Launch Screen

• Action: Default - Launch from Website
• Instance: Default is recommend - p3.2xlarge is the smallest Nvidia-RAPIDS-compatible GPU
• Security group: We recommend "Create new based on Seller Settings"
• Key pair: Use existing if you have a copy of the files, else create new (and save), and likely  `chmod 400 myfile.pem` (or chmod 644 myfile.pem)
• Launch

• Wait till "Status Checks: 2/2 checks"
• Follow the public IP / DNS link from your EC2 console to start your first Graphistry session!

 

Login & launch your first visualization & notebook

 

 

 

1. Login

• In a browser, go to the public DNS provided by the EC2 console
• Upon starting the system, you need to wait ~5 minutes, and until then, may see 5XX errors or otherwise incomplete page loads
• As a first-time user, create the initial admin account from the webpage, and be sure to record the email/user/pwd

 

2. Open a notebook

• "Analyst Getting Started" -> "See Notebook"
• In sequence, run each code cell by clicking on it and entering "shift-enter"
• On cell "g.plot()", on the upper right of the graph where it says "workbook", click the left-most icon for "Open in a new window", and click the Graphistry logo to begin
• Follow the UI guide on how to interact with the notebook

 

3. Try more notebooks

• Try other notebooks: for developers, CSV upload of your own data, various DBs, ...
• Make a copy of a notebook you want to use as a template 
• You can always edit any notebook cell, re-run it, and save your changes/outputs
• Best practice is to make sure you can "run-all" in a notebook and it gives the full results

 

Recommended additional config & troubleshooting:

See the Graphistry AWS Marketplace FAQ:

• Errors: Handle common AWS issues like no available GPUs
• IP/Domain: Set an elastic IP or domain to have a stable URL across restarts
• Enable Graphistry notebooks to communicate with secure databases
  • If `pip install` does not have your package, SSH into the instance, and then the Docker container as root
  • Make sure the Graphistry IP is permitted by the security group for your DB
• Create accounts for your users
• Setup TLS and restrict server access to your VPN

Next steps & further reading

• UI guide 
• Graphistry AWS Marketplace FAQ 
• Admin docs
• Python notebooks

 

With our big 2.0 release, Graphistry's tech and features have combined into a fast and easy way to connect to your team's data,  visually understand most large or complex problems, and do it all safely & privately. For the first time, you can now easily experience Graphistry from your own AWS account. Read on for what problems our early users are solving, how Graphistry 2.0 is helping them, and quickly getting started in AWS on your own data.