Grid CC Real-time Instrumentations Grids A real-time interactive

Grid. CC: Real-time Instrumentations Grids A real-time interactive GRID to integrate instruments, computational and information resources widely spread on a fast WAN Francesco Lelli Istituto Nazionale di Fisica Nucleare Laboratori Nazionali di Legnaro, Legnaro Italy F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

Overview • The Grid. CC Project: Introduction • Bringing Instrument into the Grid: the Instrument Element • • Instrumentation Fast Instrument Communication Channel Standard Grid Interaction Current Implementation performance analysis • The Grid. CC Test-bed: Pilot application F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

General on the Grid. CC Project • Funded by EU in the Frame Program 6 • 10 Partners from 3 EU Countries + (Israel) • About 40 people engagged Country Istituto Nazionale di Fisica Nucleare Italy Institute Of Accelerating Systems and Applications Greece Brunel University UK Consorzio Interuniversitario per Telecomunicazioni Italy Sincrotrone Trieste S. C. P. A • It is a 3 years project. Started the 1 st September 04 Participant name Italy IBM (Haifa Research Lab) Israel Imperial College of Science, Technology & Medicine UK Istituto di Metodologie per l’Analisi ambientale – Consiglio Nazionale delle Ricerche Italy Universita degli Studi di Udine • www. gridcc. org F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 Italy Greek Research and Technology Network S. A. Greece

The Grid Technologies to extend the limit of a single computer (center) Storage Element Computing Element Grid Gateway Grid Technologies User Interface Computing Element F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 Computing Element

Extending the Grid Concepts Grid Gateway Terrestrial probes to monitor The volcano activities Satellite views to monitor the volcano Grid Technologie s Control and Monitor Room F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 To model calculations and disaster predictions

The Grid. CC Project ulations lc odel Ca ta for M Da ns Predictio Instruments Grid + Computational Grid. CC Project F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

Instrument Element: global scenario Instrument Element Virtual Control Room Computing Element Web Service Interface User direct Action Exec. Wf. MS Service Indirect Action F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 WMS Storage Elements Element Agr. S Existing Grid Infrastructures

The Grid. CC Architecture Virtual Control Room (VCR) All end user access is via Collaborative the VCR The IE is a Services (CS) Users generally virtualization of not working alone Information the real physical and Monitoring “Fast” all pervasive instrument Services messaging system Direct access to IE (IMS) Instrument SE (and CE) Services elements Execution possible Information Instrument System Slowly updating often not desirable but (IE) elements (IS) Compute information and Of course Instrument (IE) Storage Elements Of course there may be Security is elements Watching (via the IMS) More complex (with many IEs essential to the Security advanced Many CEs (IE) forworkflows, anywhere problems including Services reservation) successand SEs of the in the system and advanced project acting to resolve them. reservation and Qo. S Compute Storage guarantees , allowed element Virtual Control Room (VCR) Element (SE) Storage (CE) Element (SE) Compute element (CE) Storage Element (SE) F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 Compute element (CE) Global Problem Solver

IE Requirements Web Services Storage Element Computing Element Instrument Element Any Protocol or physical connection Sensor Instrument Network Instrument Computing Element D F A Instrument Element 1: Provide a uniform access to the physical device W E Grid C B F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 2: Allow a standard grid access to the instruments 3: Allow the cooperation between different instruments that belong to different VOs

Instrument Element: a Black Box Quick Answers to the previous slide: 1) The VIGS provide the a uniform instrumentation way 2) The fast communication channel Grid disseminate the acquired information Interaction between instruments 3) The Data Mover provide a standard Grid Interface in order to be accessed by others Grids components like the SE and the CE Instrumentation Instruments IE Data Mover Instrument VIGS Fast communication channel • The term Instrument Element describes a set of services that provide the needed interface and implementation that enables the remote control and monitoring of physical instruments. IE Key Developers: E. Frizziero 1, M. Gulmini 1, 3, F. Lelli 1, 2 , G. Maron 1, A. Oh 3, A. Petrucci 1, S. Squizzato 1, S. Traldi 1 1 Istituto Nazionale di Fisica Nucleare, Laboratori Nazionali di Legnaro 2 Dipartimento di Informatica, Università Ca’ Foscari di Venezia 3 CERN European Organization for Nuclear Research F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

Instrumentation F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

Device Virtualization Model 1. 2. 3. 4. Instrument Parameters hold configuration information Attributes hold instrument variables Control Model hold actions XML Based Language to allow the device to describe itself Attributes Control Model Voltmeter XML Based Language • Parameters: Maximum Voltage, Minimum voltage • Attributes: measured Voltage • Commands: Perform a measure F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

Instrumentation lock. Instruments unlok. Instruments retrieve. Loked get. Istance get/Set Parameters get. Commands execute. Command get. State. Machine VIGS get. Contexts get. Instrument. Managers get. Info IE Instruments Crucial non-Functional Requirements: • Instruments could be order of 106 • Only authorized people should access to the instruments of a VO • The instrumentation is not a batch process like a job submission! Interactivity is mandatory get. Remote. Execution. Time get. One. Way. Cost get. Total. Method. Execution. Time • A Distribute and hierarchic implementation is mandatory • the Security overhead should be negligible We can divide the Instrumentation in 3 main parts: • The direct access to the Instruments • The advance instrument reservation (interaction with the Agreement Service (AS)) in order to achieve (hard) guarantees • The Possibility to predict the execution time of the instrumentation methods in a concurrent access (soft guarantees) F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 Instrumentation method Documentation http: //sadgw. lnl. infn. it: 2002/IEFacade

Instrument Element Architecture create() destroy() execute() get. State() Access Control Manager Virtual Instrument Grid Service (VIGS) Instrument Element Inf & Mon Service Resource Service Instrument Manager Problem Solver • The term Instrument Element describes a set of services that provide the needed interface and implementation that enables the remote control and monitoring of physical instruments. Data Mover Data Flow State Flow Error Flow Monitor Flow Control Flow Data Collector IMS Proxy Control Manager Event Processor Input Manager Real Instruments F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 FSM Engine Resource Proxy

Access Control Manager Instrument Element Implementations The IE components are typically implemented into a fully equipped Machines (e. g. dual core cpus, large memory, large disks, etc). Instrument Element Resource Service Inf & Mon Service Instrument Manager This is true for RS, IMS and PS. For IM (and DM) there are 2 possibilities, according to the application type: • IM implemented in a fully equipped machine • IM embedded into the instrument that should be controlled Problem Solver Data Mover IMS RS IM IM Embedded Web Service F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

Instrument Manager Customizable Control Manager Event Processor FSM Engine Input Manager Resource Proxy IMS Proxy Plug-in modules to interface to the instrumen Data Collector Data Flow State Flow Instruments Monitor Flow Control Flow Error Flow IM is composed by 3 main components: - Control Manager: - Input Manager. It handles all the input events of the IM. These includes commands from G errors/state/log/monitor messages. - Event Processor. It handles all the incoming message and decide where to send them. It h - FSM. A finite state machine is implemented - Resource Proxy. It handles all the outgoing connections with the resources. - Data Collector. It get data from the controlled instruments and make them available to the data mover. A is even foreseen. - IMS Proxy. It receives error/state/log/monitor information from the controlled resources and forward them F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

Partition/Configuration retrieve methods Partition and Lock setting methods Discovery Manager Subscribe Manager Configuration setting methods Discovery methods • • Partition&Lock Configuration Manager Available Resources Partition Definitions Manager Configuration Definitions RS Data Bases Resource Service Architecture The Resource Service (RS) handles all the resources of an IE and manages their partition (if any). A resource can be any hardware or software component involved in the IE (instruments, Instrument Managers, IMS components) RS stores the configuration data of the resources and download them to resource target when necessary Resources can be discovered, allocated and queried. It is the responsibility of the RS to check resource availability and contention with other active partitions when a resource is allocated for use. A periodic scan of the registered resources keeps the configuration database up to date. RS is interfaced to the WMS F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

• Instrument Manager • Instruments Instrument Manager SUBSCRIBERS Errors Log info Monitor State PUBLISHERS (Instruments nodes) Instruments Instrument Manager Information and Monitor System (IMS) The Information and Monitor Service (IMS) collects messages and monitor data coming from GRID resources and supporting services and stores them in a database. There are several types of messages collected from the sub-systems. The messages are catalogued according to their type, severity level and timestamp. Data can be provided in numeric formats, histograms, tables and other forms. The IMS collects and organizes the incoming information in a database and publishes it to subscribers. These subscribers can register for specific messages categorized by a number of selection criteria, such as timestamp, information source and severity level. F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

Problem Solver Step 3 On-line information can be analyzed in order to detect possible malfunctions On Line Analisys Data Mining Tools Algorithms evaluations : Rule Induction, Tree, Functions, Lazy, Clusters and Associative Pub/Sub Step 1 The control manager can perform an autonomous recovery action where the cost for the determination it is not so heavy. Instrument Manager IMS Proxy DB State Flow Error Flow Monitor Flow Instrument Manager IMS Proxy Control Manager Step 2 Persistent information can be analyzed in order to extract knowledge F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 Instrument Manager IMS Proxy Control Manager

Poviding Qo. S over Web Sevices Performing a remote method Invocation in a given amount of time: t 0 t 8 Processing t 7 Serialization Deserialization Client side Crucial Times are: t 3 -t 0 One Way Cost • • t 1 t 6 Transmission t 2 t 5 Deserialization Serialization Network t 4 -t 0 Remote Execution Cost F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 t 4 Service side t 7 -t 0 Total Method Execution Cost Avg =f(Cpu, Inputsize, Outputsize, Algorithm, Key-Factor, net) SDev =F(Cpu, Inputsize, Outputsize, Algorithm, Key-Factor, net) Cpu = machine HD + machine load (client and server side) Algorithm = method semantic Net = bandwidth + RTT Key-Factor = input value that change the method semantic Inputsize, Outputsize =effective type and dimension t 3 Operation execution

Virtualization of Real devices Each IM Represent the virtualization of a device Web Cam Position Max Value min Value Video Streaming linked Temperature Unlinked IE create() IM Cam IM Sensor destroy() execute() get. State() Data for Model Calculations Resource Service Inf & Mon Service F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 Data Mover Predictions Unlinked

Virtualization of Real devices (I) Each IM Represent the virtualization of a device Web Cam Position Max Value min Value Video Streaming linked Temperature Unlinked IE linked IM Cam IM Sensor create() destroy() execute() get. State() IM Master Controller Resource Service Inf & Mon Service F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 Data for Model Calculations Data Mover Predictions Unlinked

Virtualization of Real devices (II) Web Cam Position Video Streaming linked Max Value min Value Temperature Unlinked IM Cam RS IMS IE Cam Data Mover IE Sensor IMSensor Each Instrument is virtualized and a 3° IE use this others IE in order to accomplish a complex functionality IM Master Controller RS F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 IMS RS IMS IE Master Data Mover Data for Model Calculations Predictions Unlinked

Virtualization of Real devices (III) Web Cam Position Max Value min Value Video Streaming linked Temperature Unlinked IE linked Sensor Proxy create() Cam Proxy destroy() IM Master Controller execute() get. State() Data for Model Calculations Resource Service Inf & Mon Service F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 Data Mover Predictions Unlinked

Fast Instrument Communication Channel F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

Message Oriented Middleware • Topic A • Topic B • Subscribers Subscribe to a given Topic/Queue with a subscribe condition Publisher publish message in asynchronous in a given Topic/Queue way with a given message condition Publisher and subscribers can be part of the same program or in WAN distributed machines Queue Q • In Our Case: • Each instrument can be a data publisher or a data consumer • For more demanding application an instrument must send/receive data in a streaming way F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 • JMS Provide a standard set of API that standardize this communication system Many Commercial and academic implementation of this API exist in both C/C++ and Java (Narada. Brokering, Sun, IBM, Sonic. MQ etc )

RMM-JMS • • • RMM-JMS is a JMS implementation on top of our high performance Reliable Multicast Messaging (RMM) layer which provides one-to-one, one-to-many data delivery or many-to-many data exchange, in a message-oriented middleware point-to-point or publish/subscribe fashion The exceptional performance supports remote and distributed control and operation of scientific instruments such as sensors and probes Multicast transport for publish/subscribe messaging: Supporting the JMS Topic-based messaging and API, with matching done at the IP multicast level. The transport is a Nack-based reliable multicast protocol. Direct (broker- less) unicast for point-to-point messaging: JMS Queues are implemented over RMM queues. The transport is the TCP protocol. Brokered unicast transport for publish/subscribe messaging. The broker receives messages from the producer in either unicast or multicast delivery mode, and sends the messages to the subscribers in either mode broker serves as a bridge in a LAN-WAN-LAN configuration Main Contribution of IBM Haifa Research Lab (Israel) F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

Performance: message rate – the many-to-one • Blade center with 12 CPUs and 1 GB Ethernet switch • No message loss • Total throughput: 61 MBytes/sec. and 67 MBytes/sec. for (a) and (b) respectively (a) rate - msg size 1000 bytes (b) rate - msg size 100000 bytes 90000 900 800 70000 700 min 60000 Max 50000 Avg 40000 SDev msg/sec 100000 600 min Max Avg SDev 500 400 300 20000 200 100 0 0 0 5 10 15 Number of Publishers F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 20 0 5 10 15 Number of Publishers 20

Performance: message rate – the one-to-many • Blade center with 12 CPUs and 1 GB Ethernet switch • No message loss • Peak result of over than 400000 msg/sec. was reached Rate, msg size 1 Byte Rate, msg size 1000 bytes 600000 90000 80000 500000 70000 msg/sec 400000 300000 200000 60000 min 50000 Max Avg 40000 Avg SDev 30000 20000 100000 0 5 10 15 20 25 Number of Subscribers F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 30 0 5 10 15 20 25 Number of Subscribers 30

Performance: round trip time (RTT, Latency) • Two machines with a single publisher and a single subscriber on each one • Average round trip time computed over 1000 samples RTT 100 Time (m. Sec) 10 Avg Sdev Ping 1 0. 01 1 10 100 F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 10000 Messages Size 1000000

Standard Grid Interaction F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

Data Mover IE Instrument Resources Data Mover Data Collector IM Web Service Interface: get_data() SRM interface Http Server and TCP/IP raw socket IM IM • • • The task of this element is to get data from the “data collector” of the IM Data can be accessed via: – Web service interface for generic data dump (e. g. slow storage, spy stream, etc. ) – grid storage element (SE) and available CEs can access to the data via an SRM Interface – Http server and TCP communication for high performance had-hoc data transfer The Data Mover exposes its methods to the IE web service and can be instrumented itself as an instrument. F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

Current IE Implementation a fist taste F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

Instrument Manager Performances (I) F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

Instrument Manager Performances (II) 1+2 1 2 3 IM with CMS Instruments 1 3 Optimized environment F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

DB IMS Performances Pub/Sub (JMS) TCP/IP Web Service Interface IMS Proxy …. IMS Proxy Errors/log/states messages (xml and java objs) F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

Main IE Pilot Applications: Power Grid Virtual Control Room Instrument Manager Power Grid V. O Gas . . . Instrument Element Sola r F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

Main Grid. CC Pilot Applications: Control and Monitor of high energy experiments F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

Main Grid. CC Pilot Applications: Control and Monitor of high energy experiments F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

The CMS Data Acquisition 2 107 electronics channels 40 MHz • O(104 ) distributed Objects to – control – configure – monitor • On-line diagnostics and problem solving capability • Highly interactive system (human reaction time fraction of second) 100 Hz F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 • World Wide distributed monitor and control

CMS Prototype F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

CMS Prototype: IEs at work - Grid. CC middleware used for CMS MTCC (Magnet Test and Cosmic Challenge) CMS Instrument Elements TOP Det 1 Detector GTPe - 11 Instrument Elements with a hierarchical topology - Instruments are in these case Linux hosts where the cms on-line software is running 8 1 DAQ IE Instrument Managers DAQ - More than 100 controlled hosts Trigger - 25 days to the start of the data taking ! TTS Filter. Farm Fed. Builder F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 Ru. Builder

IDS Intrusion Detection System 1. Taking Control Pirated machines Domain A Target domain "zombies" X Pirated machines Domain B F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

IDS Intrusion Detection System A DDo. S Attack Domain-wise Sources of the attack Sensor Instrument Element F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 Sensor Instrument Element Target Domain

Main Grid. CC Pilot Applications: Remote Operation of an Accelerator Elettra Synchrotron F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

The other Grid. CC pilot applications • Meteorology (Ensemble Limited Area Forecasting) • Device Farm for the Support of Cooperative Distributed Measurements in Telecommunications and Networking Laboratories • Geo-hazards: Remote Operation of Geophysical Monitoring Network (see first slides) • Medical Devices need a close loop between the data acquisition and the output result F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

Conclusion • The Grid. CC project is integrating instrument into traditional computational/storage Grids. • IEs need an high interaction and interactivity between itself and the users. • The Grid. CC IE implementation is currently installed in heterogeneous applications F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

Question? • Thx for your time More information: www. gridcc. org On-line Demo at: http: //sadgw. lnl. infn. it: 2002/IEFacade Acknowledgement: The Grid. CC project is supported under EU FP 6 contract 511382. F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

Another Grid. CC applications: Migraine Attacks Treatments EEC 1. Data taking 2. Data Processing GRID 3. Result Visualization and control 1 minute loop F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 4. Action

The control of the CMS Data Acquisition Virtual Cntr. Room Supporting Services Storage Element Drift Tube CMS Subdetector Diagnostics Virtual Cntr. Room • Acquire data from a CMS Muon chamber • Move data to a Storage Element F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 Computing Element • Submit an analysis job • Retrieve the job result

The control of the CMS Data Acquisition CMS Control Structure ISCHIA Web Service Comunication Retrieve the Configuration Run Control F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 Legnaro

A New Messages Oriented Middleware F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006

General on the Grid. CC Project • Funded by EU in the Frame Program 6 • 10 Partners from 3 EU Countries + (Israel) • About 40 people engagged Country Istituto Nazionale di Fisica Nucleare Italy Institute Of Accelerating Systems and Applications Greece Brunel University UK Consorzio Interuniversitario per Telecomunicazioni Italy Sincrotrone Trieste S. C. P. A • It is a 3 years project. Started the 1 st September 04 Participant name Italy IBM (Haifa Research Lab) Israel Imperial College of Science, Technology & Medicine UK Istituto di Metodologie per l’Analisi ambientale – Consiglio Nazionale delle Ricerche Italy Universita degli Studi di Udine • www. gridcc. org F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 Italy Greek Research and Technology Network S. A. Greece

Grid. CC Main Architecture Security. Aut. S Service Instrument Element TGS Pol. R Computing Element Virtual Control Room Web Service Interface Storage Elements Element Exec. Wf. MS Service WMS Agr. S User direct Action Indirect Action F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 Global Problem Solver Existing Grid Infrastructures

Instrument Element Facade Virtual Instrument Grid Service (VIGS) move. To. SE read submit. Job Logs, Errors, States, Monitors IMS IM IM IM Commands Status, Parameters Data Mover Grid Operations IE Grid. FTP Move. Data Grid Operations get. Istance get/Set Parameters get. Commands execute. Command get. State. Machine RS Instrument Element VIGS get. Contexts get. Instrument. Managers get. Info Submit Job to Grid Fast Output Channel IMS Subscribe (JMS) DB F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 • The term Instrument Element describes a set of services that provide the needed interface and implementation that enables the remote control and monitoring of physical instruments.

Interacting with Instrument Elements y rit cu ce Se ervi S 1) Full Grid. CC Environment nt ent umeet rumenntt t ns e Instrumment Instr llemen E m I Ele E IMS Comp u CCmmuutiting o o p ptinng g E Eleme n Elelemnnt t me et GPS Sto rr El ltooar ge EEeem aage lem ee e m gn en tss nt t y rit cu ce Se ervi S 2) Partial Grid. CC Environment Ex Se ecuti rvi on ce VCR This mode of operation can be used when the application does not need to access CEs and SEs. It coud for instance exploit the workflow manager of the execution service to do unattended cycles of operations and control the system via VCR Ex Se ecuti rvi on ce Security Service R VC t en n umentt tr m e t Insrulm en t t u em Insr. Eem en t l Ins E m e El -I S C l. C ++ Pe su a Vi su al Ba sic ET W . N Vi F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006 JSP rl Ja va ++ t ent men umentntt tr e ns ue Instrumment Instr llemen E m I Ele E Comp u CCmmuutiting o o p ptinng g E Eleme n Elelemnnt t me et VCR 3) Standalone Environment IE is web service based, any web service compliant clients can reach it. This mode of working is very useful for small systems and for prototyping and debug large systems IMS GPS Sto rr El ltooar ge EEeem aage lem ee e m gn en tss nt t

Algorithm and Key-Factor Example • Remote method Y=F(X) where Y, X are double and F= y= -1 if x<0 y=sqr(x) if x>0 The complexity (i. e. the algorithm that need to be remotely executed) depend on the key factor X F. Lelli, Summer School on Grid Computing, Ischia, July 20, 2006