 |
I am normally allergic to trade shows and so-called professional conferences. I find the self-serving parade of vendors and industry experts to be waste of my time. Maybe this is why unconferences run by users are getting popular.
|
The IDC IT Security Roadshow 2007 held in Gan Oranim in the Tel Aviv Fairgrounds today was mostly an exception to my rule. The sessions were interesting and the networking was not bad. Kudos to Dan Yachin and the rest of the team at IDC Israel for a job well-done.
Gideon Lopez (MD of IDC Israel) opened the day with a few remarks followed by Gil Schwed who gave the keynote. Here are some notes I took during their talks:
What are the trends in IT Security?
Gideon had 6 comments:
1. There is room for small startups that provide innovation
2. Security is 4% of IT budget (not very much)
3. Companies want consolidation of security management systems
4. There is now an increased awareness of Insider threats, although we need to get past the false positive problem and deal with Extrusion prevention that gets in the way of employees
5. Security as an enabler
6. The odds between attackers and defenders is not equal and it's interesting to compare. Look at the startup community in Israel that are consuming sizable RnD budgets versus modest hacker budgets.
I thought Gideon's comments were mostly ok but "Security as an enabler"?? Give me a break. Security is not a business enabler. It is a cost. It is a way of protecting your assets so you can operate your business, get home safely at the end of the day and deliver positive returns for your shareholders.
The last point that Gideon made is interesting, because it it's meaningless.
There are attackers (organized crime) with huge budgets that can bribe trusted insiders, employ advanced technology, human honey pots and long term social engineering to steal valuable digital assets. Then there are low-budget attackers who smash and grab (like the thieves who broke into BMC Software offices in Ramat Hachayal by smashing a ground-floor window and carting out 50 workstations while the guards were on the other side of the building. Both attackers and owners of valuable assets perform an economic assessment of how much they want to spend on an attack (or defense) versus the value of the assets at risk.
This lack of understanding of the economic dimension of risk was a common denominator later in the conference in the user focus session run by Gadi Gilon, the CIO of Orange Israel.
Gil Schwed - Growth in Internet, growth in threats
Gil made the point that attacks on assets have grown proportionately to the growth the number of Web sites. Unfortunately, no one took Gil up on this seemingly trivial point, because the growth in number and variety of attacks should actually be a function of the number of people connected to the Internet, not the number of web sites. This is a far bigger number afik.
Gil's presentation was a march of time Powerpoint and a not so subtle pitch for Checkpoint.
Gil did make one excellent point, namely that customers need strong vendor security focus. Because of merger and acquisition activity (Symantec buying Veritas (a storage vendor), EMC buying RSA (a security vendor) ) etc.
Gil is correct when he says that a lot of the big vendors don't have a strong security focus anymore. The Symantec VP Business Development for EMEA (William Beer) that spoke with me at the break, confirmed that this is definitely true for his company. I guess Cisco never had a security focus.
Checkpoint is one of the few pure-play security vendors that service all sized customer segments. The distribution of the size of their customers is:
- 30% > 10,000 employees
- 30% 1,000-10,000
- 30% 50-1,000
- 10% < 50 employees
Here are some notes from Gil's well-delivered (although somewhat limited-vision) ppresentation:
Enterprise security is composed of Infrastructure, data, mobile and endpoint Security
Network security is composed of a core + data, endpoint, IAM, AA, Threat mgmt / VA, SIM
Today - network security is complicated, hard to manage, too many vendors, it is siloed between departments, reactive and inconsistent.
We need vendor consolidation. IT sec managers dont have enough time to hear, good idea with potential and a new innovative product cannot justify itself because of the market resources required to get customers attention
I suggest that we need to reduce the number of security vendors installed in an enterprise from 15 to 4 or 5.
There has been a loss of security focus for some vendors. This is because of the M&A activity in the networking vendors space. Companies get lost in the supermarket of a big vendor
There has also been a security vendors shift to other larger application spaces like storage
We need architected solutions that unify management and create interoperability, for example an integrated security gateway with FW, IPS, VPN,extrusion prevention, Virtualization and central management.
Data security challenges
1. extrusion/information leakage
2. big files, removable devices (USB, iPOD)
3. lost / stolen notebooks (60% of information theft)
Data security creates huge exposure since ccompanies required to disclose incidents, notify entities at risk. The first requirement is a need for policies on data usage in the firm, then they need port control, media encryption, and gateway protection (extrusion prevention).
Mobile client+data/network gatway +total endpoint + mgmt
Similar to general endpoint requirements
Endpoint
The endpoint requirements are also AV,FW/VPN, data, unified and managed security. We built a personal fw business from 0 to 50M (not that we had much competition) on basis of the lackings of the WindowsXP firewall
Consolidating Security Session
The participants were Blair Semple (Decrue/Netapp), Shlomo Touboul(Yoggie), William Beer (Symantec European Security Practice), Edouard Lorrain (Business development manager, Citrix Europe) The moderator was Peter Stremus (VP Biz Dev IBM/ISS EMEA)
Here are some highlights (note how the vendors don't answer the question but use it as an opportunity to tout their wares)
Can Security vendors survive as pure play, or must they be acquired?
Citrix - security is a monitoring layer that is part of the application architecture
Symantec - security intelligence enables customers to be proactive
Yoggie - Yes, there is consolidation but when a big company grows by acquisition, they become more concerned with customer relationship management than with innovation, which is why there will always be startups.
Can we rely upon and trust the best-of-breed security vendors for our technical countermeasures?
Symantec - Takup on our MSS has been slow, because the IT security people dont want to relinquish control. We're selling co-sourcing these days
Decrue - MSS needs to be viewed as another security countermeasure to the organization vulnerabilities. You need to weigh what's best in terms of the business needs. Look - the term itself - best-of-breed is a dog-show term meaning, expensive and not integrated with anything else and high-maintenance. Make your own conclusions about best-of-breed in that context.
Yoggie - I'm both pragmatic and paranoid when it comes to trusting security technology
Citrix - It's meaningful that the telecom service providers have acquired managed security services and system integrators, (BT and Counterpane, Belgacom and Telindus). I think that a key selling point for security is integration and management which the service providers are good at. The downside is that telecoms are slow moving and non-innovative. Their SME customers usually trust them but when there is a cost, performance or security issue, customers will flee the coop to a competitor. Look at the case when the UK ISP (Tiscali) had a DDOS attack and their mail servers went down - they suggested to their customers to use a free Webmail service.
Security User Experiences Session
The participants were Avi Weissman (See security), Itay Janovsky (ZIM), Itzik Kochav (Clalit), Rachel Jacoby (Bank of Israel). The moderator was Gadi Gilon(CIO Orange Israel)
What is the most important thing an Infosec manager should do?
Avi - map and valuate your assets, none of my customers do that.
ZIM - Not technology, the cost of maintaining security systems is the main issue since the cost of maintenance is much higher than the cost of acquisition
Clalit - The CISO should be a policy-setter, security should be part of the design so that it doesn't interfere with the operation It should not be a cover
When was the last time you did a risk assessment and did you calculate economic values of risk?
"We generally ignore economic value of risk and we are shooting ourselves in the foot when we don't evaluate risk in financial terms."
Note how everyone talks about what should be done without admitting guilt.
Gadi - (asking for a show of hands in the audience) - almost no one raised their hand
Clalit - We do it annually (or did he mean, should do annually, I'm not sure I heard right...)
Avi - You're right, most customers don't do it but they should. There are far more technical countermeasures than threats so it must be an economics decision - the first part of an quantitative risk assessment is identifying and then valuating the assets.
Rachely Jacob(Bank of Israel) - The 357 Infosec standard mandates use of probabilities of occurence (ARO), of course a bank that complies with 357 and Basel II can use the AMA and mitigate risk while allocating resources to different countermeasures all on a fiscal basis.
ZIM - Risk management is a business process, how much security is enough for the organization, I would say that we need to breakdown the question of quantitative risk assessment into 4 areas: BCP, Baseline (80/20 rule for countermeasures effectiveness),
Basic risk management practices for systems and awareness for employees.
Won't regulation create more vulnerabilties because of it's cost and checklist mentality?
Rachel (BOI) 357 has been effectiveness for IT governance in Israel, even though IT security interferes with implementation projects
Avi - Better to employ countermeasures dictated by a compliance standard than to do nothing. Israel needs regulation...(I'm not sure he is living in the same country I live in - Israel has 10 different regulations for privacy compliance and none are enforced, Israel has regulation for use of the radio spectrum and because it isnt enforced, pirate radio stations cause near-plane crashes at Ben-Gurion Airport.
I would rephrase that - as Israel needs less regulation and more enforcement.
What should be the relationship between the IT manager and the information security manager?
Avi - There should be separation of duties between IT and security, since security is a separate expertise.
Rachel (BOI) - The functions do not have to be separated (357 allows the security manager to be part of the IT group) but the IT manager needs to set policies, do risk assessments and penetration testing of applications. The security process needs to integrative in the organization.
Clalit - In large organizations like ours, there needs to be three bodies: A legislative (that sets policy), an executive (that executes policy) and an audit (that monitors execution against policy)
ZIM - It depends on the corporate culture.
I'm not going to report on the other sessions
The Insider threat session featured a bunch of vendors from McAfee, Symantec, Intellinx and Websense talking about their respective endpoint or gateway or security intelligence perspectives) - I was outside having an interesting talk with William Beer from Symantec.
The Rethinking security session was pretty weak and although there were some smart people there (Dan Yachin from IDC, Yaron Polak (Genesis) Anat Bremler (formerly Riverhead) and Moshe Ishai (CTO of Comsec) Miri Hizkayev (IBM IGS) the responses did not add value for me - there were generalizations like the next big thing will be a security co-processor or quantum cryptography and a redux of the security is a business enabler canard that from Comsec.
Other than that it was a nice conference. I enjoyed seeing colleagues and running into Elisheva Jakobovich from Vertex and Jonny Saacks from Genesis Partners.
The food was pretty good too.