Chapter 19
Distributed Systems: Resilience,
Desperation
Igor Schagaev and Stephen Farrell
Abstract Major drawbacks of existing distributed systems, including networking
are discussed. To cope with them we introduce a new design concept of performance-, reliability-, energy- smart system, applied for distributed system and networking. Our approach of distributed system handling was called a desperation
control, which introduce and explain. It assumes introduction of new routing algorithms for routing and for package handling inside routers. We claim that state of
a system as a whole as well as state of a package should be considered and treated
in concert varying ways of package handlings. Shown that proposed approach at
order of magnitude boosts performance and reliability of network making it suitable
for real-time packages and widening applications for really important applications:
existing technologies and billions of devices should serve and make people better.
Using elements of graph theory, we analyze and compare proposed algorithms with
known, explaining our advantages. Weighted costs of each segment of package
path, combined with vector heuristic introduced and enriched by graph-logic
extension for each routing table makes proposed concept the best so far from all
known. Resilience of distributed computer systems as well as networking therefore
becomes improved, extending application domain to the really important applications like safety critical, military, banking, health real-time monitoring, transport
and similar.
19.1 Introduction
Computer system resilience we have introduced in [1]. But this book does not cover
one important and even booming segment of computer systems—distributed
computer systems and network computer systems. Clear, that having thousands of
manufacturers and overcrowded society of applications to argue about unified
hardware design is a kind of naïve. But following our mantra through all our works:
ICT must be PRE-smart—read: Performance, Reliability and Energy efficient and
smart. “Smartness” here means an ability to trade resources available for achieving
requested functionality in spite of all odds: read resilience of the system is
© Springer Nature Switzerland AG 2020
I. Schagaev et al., Software Design for Resilient Computer Systems,
https://doi.org/10.1007/978-3-030-21244-5_19
267
Distributed Systems: Resilience,
Desperation
Igor Schagaev and Stephen Farrell
Abstract Major drawbacks of existing distributed systems, including networking
are discussed. To cope with them we introduce a new design concept of performance-, reliability-, energy- smart system, applied for distributed system and networking. Our approach of distributed system handling was called a desperation
control, which introduce and explain. It assumes introduction of new routing algorithms for routing and for package handling inside routers. We claim that state of
a system as a whole as well as state of a package should be considered and treated
in concert varying ways of package handlings. Shown that proposed approach at
order of magnitude boosts performance and reliability of network making it suitable
for real-time packages and widening applications for really important applications:
existing technologies and billions of devices should serve and make people better.
Using elements of graph theory, we analyze and compare proposed algorithms with
known, explaining our advantages. Weighted costs of each segment of package
path, combined with vector heuristic introduced and enriched by graph-logic
extension for each routing table makes proposed concept the best so far from all
known. Resilience of distributed computer systems as well as networking therefore
becomes improved, extending application domain to the really important applications like safety critical, military, banking, health real-time monitoring, transport
and similar.
19.1 Introduction
Computer system resilience we have introduced in [1]. But this book does not cover
one important and even booming segment of computer systems—distributed
computer systems and network computer systems. Clear, that having thousands of
manufacturers and overcrowded society of applications to argue about unified
hardware design is a kind of naïve. But following our mantra through all our works:
ICT must be PRE-smart—read: Performance, Reliability and Energy efficient and
smart. “Smartness” here means an ability to trade resources available for achieving
requested functionality in spite of all odds: read resilience of the system is
© Springer Nature Switzerland AG 2020
I. Schagaev et al., Software Design for Resilient Computer Systems,
https://doi.org/10.1007/978-3-030-21244-5_19
267