Anaconda CVE curation process

High-quality, accurate, and dependable CVE information

CVE Data Source Automated Matching Human Curation Refined CVE Metadata

The National Institute of Associating NVD CVE data Anaconda staff review NVD Accurate CVE metadata
Standards and Technology with packages in the CVE data for accuracy and allows organizations to filter
(NIST) National Vulnerability Anaconda Repository then categorize, refine and out OSS packages that don’t
Database (NVD) improve the reported meet their security
information. If possible, requirements
CVEs are patched.

Confidential & Proprietary. © 2023 Anaconda 2

Holistic approach to securing the supply chain

Confidential & Proprietary. © 2023 Anaconda 3

Anaconda Data Science Platform Architecture

"Manual deployment
was painful and
complex. Anaconda
took deployment time
from 6 weeks to no more
than 6 minutes, with just
one click."

-Fortune 500 financial services

company

Confidential & Proprietary. © 2023 Anaconda 4

Develop and deploy projects in one platform
Iterate faster, collaborate by project, and deploy in one click with Anaconda’s data science platform

Develop Collaborate Govern Deploy

Access a wide selection of Build and refine models as Maintain enterprise security Sidestep the DevOps
tools to build containerized, a team in a version- and while enabling users to bottleneck and deploy
easily reproducible projects source-controlled create, deploy and share models with just one click
at scale environment custom packages

Confidential & Proprietary. © 2023 Anaconda 5

Which deployment is right for you?
Comparing Anaconda security features by implementation
Cloud On-Prem Managed

Trusted Anaconda Packages

Security policy engine

OSS License policy engine

Curated vulnerability data

Custom channels
Software Bill of Materials (SBOM)

Installed artifact repository (Server)

Third-party artifact distribution
Dedicated instance hosted by Anaconda

Confidential & Proprietary. © 2023 Anaconda 6

 Anaconda Repository

Public
Package
Other Internet Today, there is a huge
Sources
Repos
ie. GitHub risk in data science
(PyPI, CRAN)
workflows
Public Internet
Today, users:
Corporate Network
● Work in notebooks
on their laptops
● Download data on
their machines
● Download unknown
packages and use
Cloud Database with company data
● Strain available
compute resources

Today, IT admins:
● Lack visibility and
oversight of
packages, code, and
Individual Workstations data on workstations
Confidential & Proprietary. © 2023 Anaconda 7
 Security: On-Premises
Anaconda Repository Leverage a dedicated artifact
Public repository (Anaconda Server)
Other Internet
Package
Sources which enables a centralized
Repos
(PyPI, CRAN)
ie. GitHub location for all build artifacts

● Single sign-on
Public Internet authentication
● Custom channels, security
Corporate Network and licensing filters to
Single
Trusted ensure compliance
Repository ● Can be deployed on-prem,
(Server) air-gapped, in your private
cloud, or as a managed
instance hosted by
Anaconda

Individual Workstations

Confidential & Proprietary. © 2023 Anaconda 8

 Facilitate the data science workflow

Cloud

Get Data Open IDE / Explore & Train & Test Present Deploy Monitoring Business
Jupyter nb Visualize Models Findings Model / Feedback Impact
Anaconda Repository

Data Scientist
Public Package
Repositories

Data is siloed on
Challenges individual machines, Limited compute to train Deployment hurdles and
creating opportunity models on large security blockers derail
for data breaches datasets, difficult to and delay projects,
collaborate with others preventing teams from
achieving real business
impact
Work is siloed on individual Deployment is cumbersome,
machines, powered by takes days to weeks of time
potentially malicious and multiple internal resources
packages from unverified
sources

Confidential & Proprietary. © 2023 Anaconda 9

Enable deployment at scale

Cloud

Engineering

Anaconda
Repository
Operations

Community
Repositories
Finance

Business Intelligence

Data Scientists

Confidential & Proprietary. © 2023 Anaconda 10

Cyberattacks in the headlines

Confidential & Proprietary. © 2023 Anaconda 11