Juniper Apstra Datasheet
Download DatasheetProduct Overview
Juniper Apstra, a turnkey, multivendor automation solution, allows customers to design, build, deploy, operate, and assure data center networks, simplifying and automating data center operations.
Apstra provides:
— A singular view into the relationships and interdependencies between millions of data center elements
— Continuous real-time validation that enables you to instantly pinpoint and quickly resolve issues across all infrastructure silos
— A complete fabric management solution, regardless of vendor, for single vendor and multivendor environments
— Integration with Apstra Cloud Services for AI operations (AIOps) and application assurance
Product Description
Digital transformation is underway everywhere, and data center traffic has increased at a rapid pace. To ensure business success, you need to adapt quickly to the changes coming all around you. To achieve that, Juniper® Apstra software transforms your data center network operations by providing simplicity, reliability, and multivendor support.
Juniper Apstra is a multivendor, intent-based networking solution that provides closed-loop automation and assurance to provide a complete fabric management solution.
Apstra translates business intent and technical objectives to essential policy and device-specific configuration, and it continuously self-validates and resolves issues to assure compliance. You specify the “what” (network topology, VLANs, desired capacity, redundancy requirements, access rules, and more), and Apstra delivers the “how.”
Apstra is a lightweight, scalable virtual machine (VM) that communicates with device agents installed for each managed switch and server in the network.
You can design your rack types and fabric network using Apstra templates. Details such as single/dual-homing of servers, collapsed/3-stage/5-stage style of fabric, Ethernet VPN (EVPN)/IP fabric, and IPv4/IPv6 underlay can be specified as part of the template type and options. Once the fabric template is completed, it can be instantiated into blueprints, each representing an actual physical network. The allocation of the managed devices and network resources (“build phase”) is done within the blueprints. As the blueprint is built, Apstra automatically produces the necessary configuration for devices, providing an abstraction layer across vendors. Apstra provides continuous validation against intent and policy assurance, and it identifies configuration drift in real time, confirming that security policies are enforced as intended. Once the user commits the changes, the incremental configuration is pushed to the Juniper, Cisco, Arista, or Dell devices.
Apstra manages the entire network life cycle, giving you the ability to easily expand and scale your network and extract meaningful device telemetry and application flow data. Apstra keeps your intent in check with the actual status of the network, providing you with actionable insights into your network to ensure that your goals are met. Apstra also integrates with Apstra Cloud Services, a suite of AI-Native, cloud-based applications for AI operations (AIOps) that help you expand from network assurance to application assurance.
Features and Benefits
Apstra offers the following features:
Intent-based network design and operations
Intent-based data center automation increases application availability and reliability, simplifies deployment and operations, and dramatically reduces costs for enterprises, cloud service providers, and telco data centers. As the only intent-based networking technology to be hardware- and device OS-vendor agnostic, Apstra delivers on the vision of complete end-to-end data center automation, integrating capabilities such as group-based policies, enterprise scale, and significant intent-based analytics enhancements.
Life Cycle Management for Data Center Networks
Typically, architects design the network and operators manage it, resulting in a breakdown in information sharing and the absence of a single source of truth (SSOT). Architects are not aware of changes made to the network and operators are not fully informed of the capabilities and known limits of the system. Apstra eliminates these issues by creating an SSOT in the intent datastore and tracking all network moves, additions, and changes. Not only does Apstra track changes made to the network by other systems, but it also provides simple workflows for implementing changes across the entire network.
Advanced Telemetry: Intent-Based Analytics
For Day 2 operations, continuously monitoring and validating the operational state of your network infrastructure is critical to ensure that your intent is met, services are delivered as expected, and your network is healthy. Collecting telemetry data is necessary to achieve these goals, but not sufficient. Network operators frequently find themselves drowning in telemetry data collected by traditional monitoring systems because they don’t have the tools to analyze the data and extract actionable insights.
Apstra's intent‑based analytics (IBA) let you define the way you want to monitor and analyze telemetry data from your network. IBA enables you to create expert‑level rules and systems checks and embed them into the network management system so that they run continuously and update automatically as the network changes. With IBA, you can create user‑defined, real‑time analytics pipelines using any of the built‑in processing functions, allowing you to codify the most complex troubleshooting scenarios and automate the extraction of insights from raw telemetry data.
Apstra ships with a variety of predefined IBA probes leveraging built‑in telemetry collectors. For Juniper devices, you can also create custom collectors from any “show” command, enabling you to collect any data that is available from those devices, and then leverage IBA processing functions to apply custom analytics to the data and present the results in custom dashboards.
While IBA probes provide you with powerful real‑time analytics, you can also gain insight on historical data through analytics reports. These reports leverage statistical analysis to help you understand the distribution of your data set over time, identify outliers, and analyze trends.
Scalability in Small and Large Data Centers
Apstra was designed to handle the largest data centers in the world, supporting hundreds of thousands of connected servers. This is achieved through support for 3‑stage or 5‑stage Clos IP fabrics with EVPN‑Virtual Extensible LAN (VXLAN) deployed as the overlay. Apstra also supports smaller fabric designs. In edge data centers, for example, only a couple of switches are deployed, but the number of deployments is large and highly distributed. Apstra can easily consolidate all operations across the edge data centers into a single management interface.
Regardless of the number and scale of deployments, is focused on intent and on translating that intent to configuration. Operators can easily make changes to these roles, driving large‑scale changes to configurations across multiple vendors and network designs. To satisfy these demands, Apstra is built with a high‑throughput, highly‑scalable graph datastore that tracks all changes in real time, relieving the organization from having to manage individual IP addresses or configurations. This allows operators to focus on business‑specific needs rather than low‑level troubleshooting or reconfiguring of the network management system following every network change.
Flexible Fabric Design and Connectivity
Apstra solves the deployment and operational complexity of next‑generation data center networks by providing easy‑to‑deploy and highly‑validated fabric reference designs that can be used for any network size and cover multiple use cases. However, some data centers may require specific or out‑of‑the‑box topologies, protocols, or architectures that are not addressed by these templates.
Apstra’s “freeform reference design” allows you to build the design you want, how you want it. You can leverage any feature, protocol, or architecture that fits your deployment scenario. Freeform presents an interactive canvas to visually design or model any arbitrary network topology. The configuration is administered via configuration templates that grant you complete control over the configuration on the devices.
You can still leverage the same simple and powerful life cycle management features, including device operating system upgrades, simple device deployments, pre‑deployment data center modeling, device telemetry, analytics dashboards, powerful intent‑based analytics, and Time Voyager.
Apstra also offers flexible connectivity options for servers, firewalls, and external routers. These connectivity options can be quickly attached to any port in the fabric, with deterministic configuration to ensure that all protocols are properly functioning.
Time Voyager
A key operational feature for any network operator is rapidly recovering from human error. This is typically a complex, vendor‑specific process that requires a complete understanding of the full state of all boxes and their relationships to each other at certain points in time. The Time Voyager feature speeds up time to resolution by enabling the operator to move the entire state of the network (intent, configuration, and continuous validations) backward or forward with a few simple clicks, returning it to a specific point in time. This unique ability is enabled by its foundational intent‑based approach, including its SSOT and assurance validations.
Data Center Interconnect
As networks expand and applications require greater geographic diversity of data centers, several vendors have introduced proprietary data center interconnect (DCI) features to address stretched Layer 2 domains and active/active topologies between data centers. By contrast, Apstra supports an industry‑standard EVPN‑VXLAN overlay that extends Layer 2 application segments outside of the Apstra‑managed topology. This allows architects to integrate multiple disparate computing centers for effective load balancing, legacy migration, disaster recovery, or resource sharing while isolating failure domains for high availability and resilience. Apstra automates configuration of DCI with VXLAN tunnel stitching, dramatically reducing the complexity of unifying multiple data centers.
Access List Policies Assurance
Apstra security policy provides a simple user interface and API that allows users to define policies to control the flow of traffic between virtual networks, IP endpoints, and routing zones. The policy is automatically applied as an L3 ACL on the relevant enforcement points, radically simplifying the management and reducing the size of access control lists. Furthermore, Apstra can detect conflicts when multiple policies are applied within a blueprint overlap and automatically resolve the conflicts based on user settings such as “more specific first” or “more generic first.” Users can search existing policies based on source/destination object and by type of traffic (protocol and port number) to determine if a certain traffic flow is affected by any active policies.
Support for All Modern Network Platforms
Apstra offers the industry’s first and only vendor‑agnostic intent‑based networking platform, allowing enterprises to design a network without consideration for the hardware platforms that will eventually be deployed. The tools used to design and manage the network are the same, regardless of which vendor hardware or network operating system is ultimately selected. This translates to a massive reduction in OpEx by eliminating the need to maintain staff expertise in multiple platforms and vendor nuances. There is also an opportunity to reduce CapEx by allowing all modern vendors to be considered for inclusion in an Apstra‑managed environment.
Flow Data
Apstra Flow, a multivendor network observability solution for modern data centers, delivers unprecedented insights at any scale for network performance, availability, and security. Apstra Flow provides complete network visibility and in‑depth analysis of traffic patterns so administrators can pinpoint the cause of a problem and resolve the issues. Having this extensive visibility optimizes network performance, enhances security, and improves capacity planning and cost control.
Apstra Flow provides granular information about network traffic flows, including source and destination IP addresses, ports, protocols, and the amount of data transmitted. When network administrators have this information, they can better understand the network's performance and identify potential issues, such as congestion, high latency, or packet loss.
Network engineers can leverage the insights from Apstra Flow to implement strategies that optimize network traffic flow, reducing latency, balancing loads across network paths, and ensuring the most efficient use of available resources.
Apstra Flow’s ability to enrich multivendor network traffic with organization‑specific information enables more in‑depth analysis and a better understanding of network traffic patterns, resource usage, and security risks. Apstra Flow helps organizations improve their security posture, detect and respond to threats more effectively, and maintain compliance with regulatory requirements.
Apstra Cloud Services also uses Apstra Flow data to provide additional AI‑Native application awareness and assurance capabilities.
VMware Integrations
Apstra tightly integrates with VMware NSX‑T and VMware vCenter to provide network operators visibility into virtual workloads and networks. The built‑in validation speeds up the troubleshooting of virtual networking, port‑group/fabric VLAN/Link Aggregation Control Protocol (LACP) mismatch, and VM traffic issues. Remediation workflows help users resolve misconfiguration of VLANs faster by automatically suggesting the correct network fabric changes.
Feature | Standard | Advanced | Premium | Connector for VMware |
Network design | ||||
3-stage and 5-stage Clos design | X | X | X | |
Collapsed fabric design (Edge data centers) | X | X | X | |
L2 access switches | X | X | X | |
High Availability switches at the access layer | X | X | X | |
Freeform design (any network design) | X | X | X | |
IPv4 fabric (non-EVPN) | X | X | X | |
IPv6 fabric RFC-5549 (non-EVPN) | X | X | X | |
EVPN fabric | X | X | X | |
Virtual routing and forwarding tables (VRFs) | X | X | X | |
L2/L3 virtual networks (IPv4/IPv6) | X | X | X | |
Intra-rack (VLAN), or inter-rack (VXLAN) virtual networks | X | X | X | |
Single or dual homing of external systems (MLAG/vPC/CLAG/ ESI) | X | X | X | |
L3 sub-interfaces | X | X | X | |
Dynamic Host Configuration Protocol (DHCP) relay | X | X | X | |
External BGP peering | X | X | X | |
Dynamic BGP neighbors | X | X | X | |
Granular import/export routing policies | X | X | X | |
Static routes | X | X | X | |
Remote EVPN gateways for L2/L3 Data Center Interconnect (DCI) | X | X | ||
Integrated Interconnect/VXLAN Stitching (DCI) | X | X | ||
Mixed vendor fabrics (i.e. fabrics with non-Juniper devices) | X | |||
Device OS | ||||
Junos® operating system and vJunos-switch | X | X | X | |
Junos OS Evolved and vJunosEvolved | X | X | X | |
Cisco NX-OS and NX-OSv | X | |||
Arista EOS and vEOS | X | |||
Enterprise and Edge Standard SONiC® | X | |||
Telemetry services | ||||
Address Resolution Protocol (ARP) table | X | X | X | |
Media access control (MAC) table | X | X | X | |
BGP session | X | X | X | |
Hostname | X | X | X | |
Interface and interface counters | X | X | X | |
Transceiver information | X | X | X | |
Link aggregation group/multichassis link aggregation group (LAG/MLAG) information | X | X | X | |
Link Layer Discovery Protocol (LLDP) information | X | X | X | |
Resource utilization (disk, memory, CPU) | X | X | X | |
Device Environmental Health (power supply, fan temperature, etc.) | X | X | ||
Telemetry services health | X | X | X | |
Custom Telemetry Collector (any additional telemetry) | X | X | ||
IP Route table | X | X | X | |
Active configuration | X | X | X | |
EVPN flooding table | X | X | ||
EVPN routing table | X | X | ||
Flow data (sFlow, NetFlow, IPFIX, and IFA) | X | |||
Intent‑based analytics (IBA) | ||||
Custom dashboards and widgets | X | X | X | |
Programmable probes | X | X | X | |
Tags and property sets for custom probes | X | X | X | |
Device system health and environmental checks | X | X | X | |
Device traffic and headroom | X | X | X | |
LAG imbalance | X | X | X | |
MLAG imbalance* | X | X | X | |
ESI imbalance* | X | X | X | |
Equal-cost multipath (ECMP) imbalance for fabric interfaces | X | X | X | |
MAC monitoring | X | X | X | |
Telemetry streaming via protocol buffers* | X | X | ||
Bandwidth utilization | X | X | ||
Critical services: utilization, trending, alerting* | X | X | ||
Leafs hosting critical services: utilization, trending, alerting* | X | X | ||
Drain traffic anomaly | X | X | X | |
Equal-cost multipath (ECMP) imbalance for spine to super spine interfaces* | X | X | ||
Equal-cost multipath (ECMP) imbalance for external interfaces | X | X | ||
Spine fault tolerance* | X | X | ||
EVPN-VXLAN type-3 route validation* | X | X | ||
EVPN-VXLAN type-5 route validation* | X | X | ||
VXLAN flood list validation* | X | X | ||
EVPN host flaps detection* | X | X | ||
BGP flapping detection | X | X | ||
Hot/cold fabric ports | X | X | ||
Hot/cold spine to super spine* | X | X | ||
Hot/cold‑specific interfaces | X | X | ||
Packet discard | X | X | ||
Interface flapping | X | X | ||
Total east-west traffic* | X | X | ||
Optical transceivers | X | X | ||
Display external routes* | X | X | ||
Connectivity fault model* | X | X | ||
Cabling fault model* | X | X | ||
Multi-agent detector (Arista only)* | X | |||
Hypervisor and fabric VLAN configuration mismatch* | X | |||
VMs without fabric configured VLANs* | X | |||
Hypervisor and fabric LAG configuration mismatch* | X | |||
Hypervisor missing LLDP configuration* | X | |||
Hypervisor maximum transmission unit (MTU) mismatch* | X | |||
Hypervisor MTU check* | X | |||
Hypervisor redundancy check* | X | |||
Platform | ||||
Apstra server backup/restore | X | X | X | |
Apstra server health reporting | X | X | X | |
Apstra sever upgrades | X | X | X | |
RESTful APIs | X | X | X | |
API User Guides and API Explorer | X | X | X | |
Graph model and GraphQL/QE API | X | X | X | |
Apstra CLI | X | X | X | |
Apstra Developer SDK (Python) | X | X | X | |
Extensible on-box or off-box device agents | X | X | X | |
Multiuser administration | X | X | X | |
Role-based access control | X | X | X | |
Self-integrity check | X | X | X | |
Security | ||||
Multiuser administration | X | X | X | |
Role-based access control | X | X | X | |
LDAP authentication | X | X | X | |
TACACS+ authentication | X | X | X | |
RADIUS authentication | X | X | X | |
Active Directory authentication | X | X | X | |
HTTPS UI | X | X | X | |
Apstra server security hardening | X | X | X | |
API-driven operation | X | X | X | |
Blueprint customization | ||||
Template types and options | X | X | X | |
Connectivity templates | X | X | X | |
Configlets with granular scope (e.g., interface level) | X | X | X | |
Config templates (Freeform only) | X | X | X | |
Property sets | X | X | X | |
Tags management | X | X | X | |
Resource pool management | X | X | X | |
Day-2 rack modifications | X | X | X | |
Day-2 fabric extension | X | X | X | |
Day 2+ operations | ||||
Staged/commit workflows | X | X | X | |
Rollback network state (Time Voyager) | X | X | X | |
Add/remove generic systems | X | X | X | |
Add/update/remove racks | X | X | X | |
Add/remove pods | X | X | X | |
Network OS upgrade/downgrade | X | X | X | |
Change/add interface | X | X | X | |
Turning interface up/down | X | X | X | |
Break/form lags | X | X | X | |
Device maintenance | X | X | X | |
Device decommissioning | X | X | X | |
Device replacement | X | X | X | |
Resource utilization | X | X | X | |
Virtual network management with bulk operations | X | X | X | |
Policy assurance | ||||
Configuration drift detection | X | X | X | |
Routing Zone constraint policies | X | X | X | |
Access list policies—conflict detection and resolution | X | |||
802.1x Network Admission Control | X | |||
Traffic control with ACLs | X | |||
Policies management | X | |||
Cabling map: anti-affinity policies | X | |||
Security policy (firewall filters/access control lists) | X | |||
Universal zero touch provisioning (ZTP) with Graphical User Interface | X | X | X | |
Device agent installer | X | X | X | |
Life cycle management | X | X | X | |
Device quarantine | X | X | X | |
Device maintenance | X | X | X | |
Virtual infrastructure integration | ||||
VMware vCenter | X | |||
VMware NSX-T | X |
Ordering Information
Please contact your Juniper sales representative for information on ordering Juniper Apstra.
About Juniper Networks
Juniper Networks believes that connectivity is not the same as experiencing a great connection. Juniper's AI-Native Networking Platform is built from the ground up to leverage AI to deliver the best and most secure user experiences from the edge to the data center and cloud. Additional information can be found at Juniper Networks (www.juniper.net) or connect with Juniper on X (Twitter), LinkedIn, and Facebook.
1000699 - 012 - EN OCTOBER 2024