SCADA RELIABITY SECURITY CLASSICAL DEFINITIONS SCADA RELIABILITY The

SCADA RELIABITY & SECURITY

CLASSICAL DEFINITIONS SCADA RELIABILITY The probability that a system including all hardware, firmware & Software will satisfactorily perform the task for which it was designed or intended for a specified time & in a specified environment. SCADA SECURITY Security of a system is a combination of its ability to support System availability Data integrity – It is the measure of the quality of information. Data Confidentiality- Protection against unauthorised / unintended access

CONCEPT OF RELIABILITY & AVAILABILITY • Reliability is usually expressed in terms of MTBF (Mean Time between failures. ) • Numerically larger MTBF indicates higher reliability. TIME TO REPAIR (MTTR) Time it takes to repair a faulty equipment. A numerically short MTTR is a system design goal.

AVAILABILITY • Availability is a function of reliability & Time to repair. • Availability = MTBF +MTTR Higher Availability requires long MTBF & Short MTTR dictates the period of outage. Overall System Availability=Product of the availabilities of each Subsystem.

FACTORS TO BE CONSIDERED FOR AVAILABILITY • • • Distributed Configuration based on function – Modularity principle—Locate groups of functions as closely as possible to the data they access frequently to avoid Communication bottlenecks. Adequate Redundancy shall be built in at various levels depending on the importance. Changeover period from Main to Standby shall be very small so that no data loss takes place. Extensive Self diagnostics & trouble shooting features. Online replacement of faulty modules should not result in Spurious tripping. Proper Grounding. Maintaining optimum temp in the Control Room. Proper termination of Communication network/Computer devices / peripherals. Use of UPS for Master Stations & two separate DC sources for RTU. —Power Supply redundancy.

5. 1 • Battery Backup & Flash PROM facility for automatic restoration of system after power failure. • Components of same function shall be interchangeable • Online Backup facility either locally or remotely. • System shall be built on an open system design with distributed architecture. • Commn. Channel Redundancy • Two Commn. Methods of Operation. – Primary Backup Pairing – Parallel Operation • Primary Backup Pairing : It is easier to co-ordinate the reporting of event data from a single buffer system through the redundant paths. • Parallel Operation: They require an application process to coordinate & synchronise the data received over parallel paths

GUIDING PRINCIPLES FOR SECURITY • Protect, Detect & Respond- It is important to be able to detect possible attacks & respond in an appropriate manner in order to minimise the impacts. – Protect: Deploying specific protection measures to prevent & discourage electronic attack against process control system. – Detect: Establishing mechanisms for rapidly identifying actual or suspected electronic attacks. – Respond: Undertaking appropriate action in response to confirmed security incidents against process control systems.

GUIDING PRINCIPLES -contd • Defence in Depth: No Single Security measure itself is foolproof. Insufficient to rely on single firewall always go in for multiplier Security model like multiple firewalls with tightly configured rules, Antivirus software & intrusion detection schemes. • Backups & Recovery: Ensure effective Backup & Recovery procedures are in place. Test the integrity of backups regularly thro’ a full restore process. Save backups at on & off site locations

GUIDING PRINCIPLES-contd System Monitoring: • Monitor in real-time process control systems to identify unusual behaviour, increased amount of network activitycould be the result of Worm infection. • For protecting unauthorised access or modification to the System, Implement & Enforce a password policy that cover strength of passwords & expiration times. Passwords to be changed frequently. • Regular review of all access rights & decommissioning of old accounts. • Usage of Firewall with virus detection / Encryption of data. • Device. Lock tool Windows Registry change can lock all devices for unauthorised users from USB, CD, Floppy Drive& Bluetooth, Serial & Parallel ports depending on Time of the Day & Day of the week.

GUIDING PRINCIPLES-contd • Keep the SCADA LAN separate from the Corporate WAN. • VLANs distribute physical networks into Smaller logical networks to increase performance. • VLANs are achieved thro’ Configuration of Ethernet Switches. They isolate the traffic from other networks.

IMPLEMENTED RELIABILITY & SECURITY POLICIES IN SCADA • • Different levels of Password protection / authority. Separate VLAN for SCADA. No Connection to Corporate LAN. Dedicated routers only for Tata Distribution Automation for Indicom Services. Devicelock for locking all possible devices & ports. Anti. Virus Software which does not load the system. Software Firewall with stringent security rules. Redundancy at all levels. -Master Station, RTU & Communication.

EXPERIENCES OF SCADA GROUP • Inspite of all the guiding principles followed, somehow we are not able to get the reliability expected. • In case of redundant systems, when one system fails the other one invariably fails. • On the contrary single m/cs have worked much better – may be because redundancy calls for more load on the system. • The air conditioning is not upto the expectation at places, eg. Kalyan, Parel, Dharavi which deteriorates the system performance. – Solution could be to study the optimum benefits accrued from the redundancy thought as against the investment made.

REDUNDANCY SCHEME IN AREVA SCADA RELIABILITY & SECURITY IN PRESENT SYSTEMS • Master Station Redundancy Reliability of any system is evaluated in terms of frequency and duration of interruption. Main objective of improving reliability is to build any system configuration which will reduce number of interruptions and hours of failures.

REDUNDANCY SCHEMES ADOPTED IN AREVA SCADA In the present AREVA SCADA • System dual master concept is used to improve the reliability. In case of dual master configuration SCADA server will work independently on two different locations or PCs and failure of one / Communication link between Dharavi & local station does not affect any functionality of the system. Disadvantage: Any modifications done locally on server do not get reflected onto server at Dharavi –eg Operator events / tags put. Solution : Thin Client / Server Client philosophy. • RTU redundancy • Reliability of such systems mainly depends upon the quality and dependability on the communication system. In the present system high priority is given at different level of communication system reliability.

SCADA COMMUNICATION