Hitachi ID Management Suite Architecture

High-Availability Password Storage

Once deployed, Hitachi ID Privileged Password Manager becomes an essential part of an organization's IT infrastructure, since it alone houses privileged passwords for thousands of networked devices. An outage in Privileged Password Manager would mean that administrative access to a range of devices is interrupted -- a major outage to IT service.

Since servers occasionally break down, Privileged Password Manager supports load balancing and data replication between multiple physical servers. Any data updates written to its credential database are replicated, in real time, across all servers.

In short, Privileged Password Manager incorporates a highly available, replicated, multi-master architecture.

To provide out-of-the-box data replication, Privileged Password Manager includes a database service that replicates data between multiple instances. This service can be configured use either Oracle or Microsoft SQL Server databases as the physical storage mechanism. Hitachi ID Systems recommends one physical database instance per Privileged Password Manager server, normally on the same physical hardware as Privileged Password Manager itself.

The Privileged Password Manager data replication system makes it both simple and advisable for organizations to build a highly-available Privileged Password Manager server cluster, spanning multiple servers, with each server placed in a different physical site. Replication traffic is encrypted, authenticated, bandwidth-efficient and tolerant of latency, making it suitable for deployment over a WAN.

This multi-site, multi-master replication is configured at no additional cost, beyond that of the hardware for additional Privileged Password Manager servers, and with minimal administrative effort.

Privileged Password Manager Network Architecture Diagram (1)

Scaling to Support Thousands of Workstations

To manage privileged passwords on mobile workstations (typically laptops), Privileged Password Manager includes a service, which installs on the relevant PCs and which contacts a central server to coordinate local password changes.

This architecture has several important advantages:

The workstation service uses only HTTPS to communicate with the central server and works even when the workstation is connected behind NAT devices, firewalls or application proxies.
The workstation service does not randomize passwords unless it has established connectivity with the central privileged password management server. This avoids a situation where the central server does not know the new password value for a workstation.
Dynamic IP addresses have no impact on this architecture.
Physical relocation and long periods of detached network connectivity may delay updates to local passwords, but do not introduce a failure whereby the local administrator passwords on a workstation are unknown.

Privileged Password Manager is a component of Hitachi ID Management Suite. The following architectural description applies to the entire Hitachi ID Management Suite:

Privileged Password Manager is designed for:

Security:
Privileged Password Manager is installed on hardened servers. All sensitive data is encrypted in storage and transit. Strong authentication and access controls protect business processes.
Scalability:
Multiple Privileged Password Manager servers can be installed, using a built-in data replication facility. Workload can be distributed using any load-balancing technology (IP, DNS, etc.).
Openness:
Open standards are used for inbound integration (SOAP) and outbound communications (SOAP, SMTP, HTTP, etc.).
Flexibility:
Both the Privileged Password Manager user interface and all functionality can be customized to meet enterprise requirements.
Low TCO:
Privileged Password Manager is easy to set up and requires minimal ongoing administration.

Figure [link] illustrates the Privileged Password Manager network architecture:

Network architecture diagram (2)

Users normally access Privileged Password Manager using HTTPS from a web browser.
Multiple Privileged Password Manager servers may be load balanced using either an IP-level device (e.g., Cisco Local Director, F5 Big/IP) or simply using DNS round-robin distribution.
Native password changes on some systems may trigger transparent password synchronization. A password change interceptor DLL, library or exit may capture such changes and initiate transparent password synchronization.
Users may call an IVR (interactive voice response) system with a telephone and be authenticated either using touch-tone input of personal information or using a voice print. Authenticated users may initiate a password reset.
Privileged Password Manager connects to most target systems using their native APIs (application programming interfaces) and protocols and thus requires no software to be installed locally on those systems.
Local agents are provided and recommended for Unix servers and z/OS mainframes. Use of these agents improves transaction security, speed and concurrency.
A local agent is mandatory on RSA SecurID servers.
Where target systems are remote and communication with them is slow, insecure or both, a Privileged Password Manager proxy server may be co-located with the target system in the remote location. In this case, servers in the main Privileged Password Manager server cluster initiate fast, secure connections to the remote proxies, which decode these transactions and forward them to target systems locally, using native, slow and/or insecure protocols.
Privileged Password Manager can look up and update user profile data in an existing system, including HR databases (ODBC), directories (LDAP) and meta-directories (e.g., WMI to Microsoft ILM).
Privileged Password Manager can send e-mails to users asking them to register or to notify them of events impacting their profiles. Over 179 events can trigger e-mail notification.
Privileged Password Manager can create tickets on most common incident management systems, either recording completed activity or requesting assistance (security events, user service follow-up, etc.). Over 179 events can trigger ticket generation. Binary integrations are available for 16 help desk applications and open integration is possible using mail, ODBC, SQL and web services.